Predictors of cervical cancer screening uptake in two districts of Central Uganda.

Uganda's cervical cancer age standardized incidence rate is four times the global estimate. Although Uganda's ministry of health recommends screening for women aged 25-49 years, the screening remains low even in the most developed region (Central Uganda) of the country. This study examined the demographic, social, and economic predictors of cervical cancer screening in Central Uganda with the aim of informing targeted interventions to improve screening. The cross-sectional survey was conducted in Wakiso and Nakasongola districts in Central Uganda. A total of 845 women age 25-49 years participated in the study. Data were analyzed at bivariate and multivariate levels to examine the predictors of CC (cervical cancer) screening. Only 1 in 5 women (20.6%) had ever screened for cervical cancer. Our multivariate logistic regression model indicated that wealth index, source of information, and knowledge about CC and CC screening were significantly associated with cervical cancer screening. The odds of cervical cancer screening were higher among rich women compared with poor women [AOR = 1.93 (95%CI: 1.06-3.42), p = 0.031)], receiving information from health providers compared with radios [AOR = 4.14 (95%CI: 2.65-6.48), p<0.001, and being more knowledgeable compared with being less knowledgeable about CC and CC screening [AOR = 2.46 (95%CI: 1.49-3.37), p<0.001)]. Overall cervical cancer screening uptake in central Uganda was found to be low. The findings of the study indicate that women from a wealthy background, who had been sensitized by health workers and with high knowledge about CC and CC screening had higher odds of having ever screened compared with their counterparts. Efforts to increase uptake of screening must address disparities in access to resources and knowledge.

Background

Globally, cervical cancer ranks fourth amongst the most common types of cancer with about half and one third of a million new cases and deaths per annum respectively [1]. There is intense inequality in incidence globally since the biggest cervical cancer burden (84% of new cases and 87% of the deaths) occur in Low and Middle Income Countries (LMICs) [1]. Globally, Eastern Africa has the highest (30) and second highest (40.1) age standardized mortality and incidence rates respectively [1]. While widespread cytology-based screening in high-income countries has resulted in decreased incidence and mortality from cervical cancer, LMICs, with poor uptake of screening, have not seen similar improvements, and in some cases, incidence and mortality actually continue to rise. Cervical cancer is the most common incident and mortal cancer amongst women in Uganda [2]. Cervical cancer contributes to about 40 percent of all malignancies reported by Kampala Cancer Registry (KCR) [3]. Estimates show that in 2018, about 6413 women were newly diagnosed with and 2400 succumbed to cervical cancer [2].

It is projected that by 2025 Uganda will have about 6400 new cervical cancer cases and 4300 deaths annually [2]. Services for cervical cancer prevention are relatively available in Central Uganda [4]; For instance, one of the study district was the first to benefit from a pilot project of cervical cancer prevention intervention [5]. It is also the most developed region of the country that hosts the capital city [6]. Studies have indicated high acceptability of CC screening in Central Uganda [7], especially self-collection of vaginal samples [8]. The Uganda’s MOH target was to screen and treat 80% of eligible women aged 25–49 years by 2015 [4]. Ugandan women are screened by Visual Inspection with Acetic acid (VIA) and those who test positive with eligible precancerous lesions are treated by cryotherapy [4]. Screening in Uganda is unpredictable, opportunistic, and in some instances absent due to a shortage of resources or absence of will to commit financial resources [9]. This has resulted into low uptake [10] surprisingly, even in the most developed region (Central) of the country [11].

While comprehensive vaccination is cost-effective and lifesaving, incidence of cervical cancer is not expected to drop for at least 2 decades after widespread vaccination uptake [13] and in the meantime cervical dysplasia and early cancers will remain common and treatable through effective utilization of cervical cancer screening. The success of screening depends on access, uptake, and follow through the cascade to treatment for those who screen positive. According to World Health Organization (WHO), women aged 30 years should start screening for cervical cancer [14]. Additionally, the WHO guidelines recommend that screening at least once in a lifetime is beneficial, and intervals may depend on existing infrastructure and resources; decisions regarding the frequency of screening and target ages are determined by costs, existing burden of disease and infrastructure, and are left to respective governments [15].

Cytology-based screening is not practical for wide-spread use in sub-Saharan Africa due to its high cost, low sensitivity, inherent need for laboratories and trained technicians and complex follow-up protocols [4]. Testing for Human papillomavirus (HPV), the causative agent in almost all cervical cancer, is recommended as the primary screening modality where feasible [7,8]. HPV DNA testing is the most objective and sensitive screening approach [9-12], and has been shown to decrease mortality from cervical cancer in low-resource settings [13]. Visual inspection with acetic acid (VIA) is an acceptable alternative where HPV testing is cost-prohibitive [7,8]. Data suggest that self-collection of HPV, has comparable sensitivity to clinician-collection and is well-tolerated by women [11,12,14,15]. A simulation model based on epidemiologic data from Uganda shows that HPV testing may be more cost-effective than VIA [16]. Multiple proven cervical cancer screening approaches exist including: visual inspection with Lugol's iodine (VILI) or acetic acid (VIA), the Papanicolaou test (Pap smear), and HPV testing [16]. In resource rich settings, the WHO recommends testing for HPV first followed by VIA to identify women who can benefit from cryotherapy [15]. However, in resource limited settings, VIA is highly recommended because of its affordability and ability to screen and treat with in a single visit [16,17].

Previous studies about cervical cancer screening in Sub-Saharan Africa focused on either urban areas, health care settings or rural areas [10,18-20]. These studies have indicated high levels of awareness among study participants about cervical cancer, its signs, symptoms and prevention [21,22], despite the low uptake of screening services. These studies further report several factors associated with cervical cancer screening, these include; age [23], social economic status [23], source of information [10], type of place of residence [24], knowledge about CC and CC screening [18,20,25]. Studies that used implementation science approach suggest that evidence on determinants of cervical cancer screening plays a significant role in informing effective interventions [26,27]. Interventions implemented with limited evidence regarding population specific predictors of cervical cancer prevention have faced challenges associated with lack of direction, negative perception, lack of scope, and limited acceptance, hence registering limited success [28,29]. Considering the scarcity of evidence on the predictors of cervical cancer screening with rural and urban representation, the objective of the current study was to report the prevalence of screening in Central Uganda and to examine for associations between predictors and successful screening in order to inform the design of future screening programs.

Methods

Study design and setting

We conducted a population-based cross-sectional survey in two of 27 districts in Central Uganda during June and July 2019. Central Uganda has 2 urban/ semi-urban districts and 25 predominantly rural districts. The two districts considered for the study include Wakiso; a peri-urban area near Kampala city and Nakasongola district; a rural district. Wakiso district is located at the outer skirts of the capital city (Kampala) with a total population of 1997418 people [30]. A number of activities in Wakiso are influenced by Kampala capital city. Nakasongola district is located 144 kilometers (kms) North of Kampala, with a total population of 181795 people [31], and the residents are mainly subsistence agriculturalists. According to 2014 National population and housing census area specific profiles, Wakiso and Nakasongola districts had 843604 and 31659 women aged 20–60 years respectively [30,31]. Cervical cancer prevention interventions have been implemented in these two districts [4]. The target population of the survey was women age 25–49 years who had lived in the area for at least six months. The 25–49 age group was considered because it is recommended by MOH for cervical cancer screening [4].

Sample size and sampling procedure

The sample size of 850 women was calculated using Kish Leslie formula [32]. The prevalence of cervical cancer screening was estimated at 50% in order to obtain the maximum possible sample size which provided more precise estimates. The calculated study sample size of 850 was based on the estimated 50% prevalence and a precision of 5% to allow a 95% interval around estimates. Since the study used cluster sampling to select a simple random of clusters, we factored in the design effect of two and a response rate of 90%. One district was randomly selected from each of the rural (25 districts) and urban/ semi-urban (2 districts) clusters of central Uganda. We randomly selected 34 villages out of the 1916 villages from the study districts; 24 out of 1582 and 10 out of 334 villages/ wards were selected from Wakiso and Nakasongola districts respectively. A big proportion of eligible women were mainly from Wakiso therefore, we interviewed 40% more women from the district. Each selected village/ ward was considered a cluster and from each cluster, we selected 25 households using systematic sampling. From each selected household, one eligible woman was selected with priority given to spouses of household heads.

Data collection procedure

We collected data using a structured pretested questionnaire containing items (questions) adapted from tools used in studies elsewhere [6,10,19,24]. For validation purposes, we pretested the questionnaire with 10 women in the neighboring community (Mukono district); 10 kilometers away from the nearest study site to avoid contamination. The piloting of tools was conducted in an area that has characteristics similar to the study area. The tool was written in English and translation into Luganda language was done by two natives conversant in both English and the local language. We used a pair of translators who were not familiar with the original version of the questionnaire to back translate into English; the two versions were compared for conceptual equivalence and harmonized. A final translation into Luganda was then performed and checked for accuracy and preservation of meanings.

The survey tool consisted of seven sections. The first section included items on the women’s demographics such as age, educational attainment, marital status, area of residence and previous health seeking behaviors. The second section contained questions on household factors such as type of house and household assets. The third section contained questions on reproduction such as number of children, contraceptive use. The fourth section had questions about CC, the fifth section had questions about cervical cancer screening, the sixth section had questions about HPV vaccination and the seventh section had husband's characteristics.

Interviews were conducted in Luganda, the major language spoken in Central Uganda. Our Research assistants (RAs) included five women from the study districts with Bachelor degrees in Social Sciences and Education. The RAs were trained for 2 days on principles of quantitative and survey research including data collection, the objectives of the research, and procedures for; sampling, interviewing and consenting. Each RA collected data from 6 to 8 participants per day for a period of 28 days. The data collected was reviewed by the Principal investigator on a daily basis to attain quality and comparability of data among RAs.

The outcome variable of the study was cervical cancer screening. Cervical cancer screening was measured in terms of whether respondents underwent any CC screening test ever; respondents were specifically asked “Have you ever been tested or examined for cervical cancer or precancer?”(No/Yes). Explanatory variables included; sociodemographic information including age, religion, place of residence, ethnicity, marital status, and parity; as well as health characteristics like use of family planning and recent visit to health facility. Other explanatory variables that were considered are; Source of information about CC screening, and distance to screening facility. Source of information about CC screening was obtained by asking women where they first got information concerning CC screening. Knowledge sections consisted of 11 and 8 items for CC and CC screening respectively. These questions examined the women’s specific knowledge about CC and CC screening. One point was given if a respondent gave one or more correct response(s). We obtained composite knowledge and mean knowledge scores. Women who obtained scores greater than the mean were considered to have high knowledge and vice-versa. Wealth index was a composite score measured by household assets. Factor scores of household assets were generated. For this study it was recoded into three quintiles: poor, middle and rich.

Data management and analysis

Two independent clerks entered data using Epidata 3.1 software (EpiData Software, Odense, Denmark). Data was synchronized and cleaned and then exported to STATA I/C version 16 for analysis [33]. Descriptive statistics in form of frequencies were generated and chi-squared tests were used to determine associations between independent variables including sociodemographic characteristics and dependent variables; being screened for CC ever. Multivariable logistic regression models were used to explore association of sociodemographic and health related predictors with the outcome (being ever screened for CC) adjusting to limit bias from confounding. Odds ratios were reported with accompanying 95% confidence intervals. The multivariable logistic regression model comprised of explanatory variables whose p-values were less than 0.05 during the Chi-square tests. An exception was made for only type of residence because of its significance in the study. Multicollinearity tests were performed.

Ethical considerations

This study was approved by Makerere University School of Social Sciences Research and Ethics Committee (MAHSSREC) and the Uganda National Council of Science and Technology (UNCST); UNCST registration number SS4848. We obtained clearance to access communities from the district and local leaders. Voluntary written informed consent was obtained from all participants, and they were assured of confidentiality. Participants were also informed of their freedom to decline participation if they chose to and or join the study and withdraw at any point without fear of retribution from the study team. After the interviews, about 5–10 minutes were allowed for participants to ask questions about cervical cancer and disease prevention.

Results

All the 850 prospective participants approached accepted to participate in the study. Five questionnaires were incomplete (missing data on any three of residence, age, gender, number of biological children and level of education) and were excluded from analysis, leaving 845 for analysis. The study results indicate that only 1 in 5 (20.6%) of the women had ever screened for cervical cancer. Among the 845 women who participated in the survey majority were aged 25–39 years (80%), married (77.6%), in the middle wealth quintile (60.4%), Christians (80.2%), and had visited a trained health worker in the last 6 months (71.6%). Over half of the study participants were having 1–3 children (57.0%), using any form of contraception (52.5%), having high knowledge about CC and CC screening (58.3%), receiving information about cervical cancer screening from health workers (27.6%), and had attained at least secondary education (57.6%). Close to 4 in 10 (38.6%) were Baganda by tribe and very few (4.5%) women were professionals.

The bivariate results (Chi square results) indicate that cervical cancer screening was significantly associated with age (p = 0.001), occupation (p<0.001), wealth index (p = 0.020), knowledge about CC and CC screening (p<0.001), parity (p = 0.002) and source of information (p<0.001). Cervical cancer screening was higher among women who were; professionals (42.5%), rich (30.4%), having high level of knowledge about CC and CC screening (26.2%), having 4 or more children (25.6) and women whose main source of information about cervical cancer screening was health workers (41.5%). Our bivariate results indicate that religious affiliation, education attainment, study site, marital status, ethnicity, age at first marriage, visiting a health worker in the last six months, and use of contraception were not significantly associated with cervical cancer screening (Table 1).

Table 1

Distribution of women by demographics, socio-economic factors, knowledge levels and cervical cancer screening status (N = 845).

Characteristic
	Ever screened n(%)	Never screened n(%)	Subtotal n(%)	P-value
Total	174 (20.6)	671 (79.4)
Age group				0.001
25–39	124 (18.3)	555 (81.7)	679 (80.4)
40–49	50 (30.1)	116 (69.9)	166 (19.6)
Religion				0.349
Christians	144 (21.2)	534 (78.7)	678 (80.2)
Muslims	30 (18.0)	137 (82.0)	167 (19.8)
Study site				0.918
Wakiso (Urban)	123 (20.5)	477 (79.5)	600 (71.0)
Nakasongola (Rural)	51 (20.8)	194 (79.2)	245 (29.0)
Education attainment				0.694
Some primary	76 (21.2)	282 (78.8)	358 (42.4)
At least some secondary	98 (20.1)	389 (79.9)	487 (57.6)
Occupation				<0.001
Professional	17 (42.5)	23 (57.5)	40 (4.5)
Other	157 (19.5)	648 (80.5)	805 (95.5)
Marital status				0.529
Married	132 (20.1)	524 (79.9)	656 (77.6)
Single/ separated/ widowed	42 (22.2)	147 (77.8)	189 (22.4)
Ethnicity				0.818
Baganda	66 (20.3)	260 (79.8)	326 (38.6)
Baluri	42 (22.2)	147 (77.8)	189 (22.4)
Others	66 (20.0)	264 (80.0)	149 (39.1)
Age at first marriage				0.702
≤18	67 (21.9)	239 (78.1)	306 (38.1)
19–34	100 (20.1)	398 (79.9)	498 (61.9)
Wealth index				0.020
Poor	40 (17.9)	183 (82.1)	223 (26.4)
Middle	100 (19.6)	410 (80.4)	510 (60.4)
Rich	34 (30.4)	78 (69.6)	112 (13.3)
Visited a health worker in last six months				0.519
No	46 (19.2)	194 (80.8)	240 (28.4)
Yes	128 (21.2)	477 (78.8)	605 (71.6)
Level of knowledge of CC and CC screening				<0.001
Low	45 (12.8)	307 (87.2)	352 (41.7)
High	129 (26.2)	364 (73.8)	493 (58.3)
Distance to screening facility (km)				0.018
≤5	73 (23.8)	234 (76.2)	307 (36.3)
6–10	57 (23.0)	191 (77.0)	248 (29.4)
≥11	44 (15.2)	246 (84.4)	290 (34.3)
Currently using contraception				0.475
No	78 (19.5)	322 (80.5)	400 (47.5)
Yes	93 (21.5)	347 (78.5)	445 (52.5)
Parity				0.002
≤3	81 (16.8)	401 (83.2)	482 (57.0)
≥4	93 (25.6)	270 (74.4)	363 (43.0)
Source of information about screening				<0.001
Radio	41 (14.5)	242 (85.5)	283 (33.5)
Health worker	93 (41.5)	131 (58.5)	224 (26.5)
Television	17 (16.2)	88 (83.8)	105 (12.4)
Others	23 (9.9)	210 (90.1)	233 (27.6)

Associations between socio-demographic and economic factors with cervical cancer screening

We used multivariable logistic regression model to examine CC screening adjusting for study site, age, occupation, wealth index, distance to the screening facility, parity, and source of information about CC screening. Rich women [AOR = 1.93 (95%CI: 1.06–3.42), p = 0.031)] had 1.93 higher odds of having ever screened for cervical cancer compared to the poor women. Women whose main source of information about cervical cancer screening were health workers [AOR = 4.14 (95%CI: 2.65–6.48), p<0.001)] had 4.14 higher odds of having ever screened for the disease compared to women whose main source of information were radios. Women who had high knowledge about CC and CC screening [AOR = 2.46 (95%CI: 1.49–3.37), p<0.001)] had 2.46 higher odds of having ever screened compared to women with low knowledge (Table 2).

Table 2

Associations between socio-demographic and economic factors with cervical cancer screening.

Characteristic	AOR (95% CI)	P.value
District
Nakasongola (Ref)	1.0
Wakiso	0.98 (0.63–1.53)	0.927
Age group
25–39 (Ref)	1.0
40–49	1.49 (0.94–2.36)	0.091
Wealth index
Poor (Ref)	1.0
Middle	1.16 (0.74–1.84)	0.520
Rich	1.90 (1.06–3.42)	0.031*
Occupation
Professionals (Ref)	1.0
Others	1.88 (0.89–3.99)	0.098
Source of information
Radio (Ref)	1.0
Health worker	4.14 (2.65–6.48)	<0.001*
Television	0.98 (0.51–1.88)	0.957
Other	0.79 (0.45–1.40)	0.424
Parity
≤3 (Ref)	1.0
≥4	1.45 (0.97–2.18)	0.069
Level of knowledge of CC and CC screening
Low (Ref)	1.0
High	2.25 (1.49–3.37)	<0.001*

Discussion

The results of our cross-sectional study reported low level of cervical cancer screening. In our study, we found that 1 in 5 (20.6%) of the women had ever screened for cervical cancer although the ministry of health’s target was to screen 80% of the women aged 25–49 years by 2015 [4], four years before the survey. However, this finding from central Uganda is higher than findings from; Eastern Uganda (4.8%) [10], and Zimbabwe (9%) [24]. Though our findings are close to findings published from Tanzania (22.6%). Most of these studies had small sample size and were rural based. Age was not significant in predicting cervical cancer screening when other explanatory variables were controlled for in a multivariable logistic regression model. The probable reason for this finding might be that screening behaviors are independent of age; with younger and older women attending screening on their personal initiative to remain healthy and health staffs’ advise respectively [34]. However, this finding is contrary to studies from elsewhere [23,35,36] as they indicate a significant influence of age on cervical cancer screening.

Receiving relevant information regarding cervical cancer and cervical cancer screening recommendation from health providers has been found to positively affect the uptake of CC screening [10,11,36]. This finding is supported by our study which found that women who received relevant information from their health providers had higher odds of having ever screened. Health workers may be essential in health messaging because they are considered knowledgeable and trustworthy. Elsewhere, women who had discussions with health care providers regarding cervical cancer expressed higher intentions to screen [11,36]. The results of our study indicate that women’s parity was not a significant predictor of cervical cancer screening. This is in agreement with findings from Kinshasa in the Democratic Republic of Congo [19] and Eastern Uganda [10]. Our finding is surprising because it is assumed that multiparous women have got higher probability of interaction with the health workers [6], these encourage those women to screen [11,36]. However, our findings are not supported by evidence from Nepal [35] and Jamaica [36] which indicated a positive influence of higher parity on cervical cancer screening. Our findings indicate that women’s wealth index was positively associated with cervical cancer screening. This is consistent with prior studies [23,24,36]. The high prevalence of cervical cancer screening among women with a high wealth index may indicate their financial ability to afford screening services in a country where health insurance is limited [37].

Women with high knowledge about CC and CC screening had screened for cervical cancer compared to their counterparts. Several other studies have found the same result [18,38]. A study that integrated community health campaign with self-administered HPV screening in Kenya achieved high uptake [38]. Alternatively, it is likely that women become knowledgeable about the service as a result of seeking cervical cancer screening. We did not find a significant association between distance to screening facility and cervical cancer screening. This may point to the influence of other factors such as inability of women to pay for the service regardless of the distance. Only a quarter (25.5%) of the women in Uganda are in the highest wealth quintile [6] and the majority of public health facilities where these women would get the services free of charge are characterized by long waiting time, and few VIA providers [39]. Our finding is in support of prior evidence which found that proximity to services did not automatically translate into utilization [10,11]. However, our finding is not supported by findings of other studies [9,40]. These studies found a significant relationship between distance to services and utilization.

Study limitations

We were not able to assess causation because of the cross-sectional nature of our study. Secondly, this study was done in mainly two districts in central Uganda. Consequently, the generalization of the study findings to other contextually different areas may be problematic. Finally, the study may have faced a problem of social desirability since the responses about cervical cancer screening were self-reported. However, probable bias was reduced by asking the women the duration since they last accessed the service. We selected the maximum possible sample size to have a fair representation of the women in the two districts of Central Uganda. Nevertheless, internal validity may have been affected by selection bias because women who were not found in their households and those who declined to participate were excluded.

Conclusion

The findings of the study indicate a significant association between, wealth index, source of information, and high knowledge about CC and CC screening with cervical cancer screening. The above findings suggest that; provider-patient health education could be increased by utilizing times when reproductive age women are already interfacing with healthcare, like pregnancy since almost all women (97%) attend at least one antenatal visit in Uganda [6]. Screening opportunities should be expanded specifically to poor women. Alternatively, investment in interventions that increase women economic empowerment will increase the women’s financial ability to afford health care.

(XLS)

Click here for additional data file.

(DOCX)

Click here for additional data file.

(DOCX)

Click here for additional data file.

10 Jun 2020

PONE-D-20-02899

Predictors of Cervical cancer screening uptake in Central Uganda

PLOS ONE

Dear Dr. Isabirye,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Your manuscript has been assessed by two reviewers, who raise partially overlapping concerns about the description of the study and methodology that should be addressed with the appropriate revisions. In addition, please ensure that you have discussed how (or whether) the questionnaire was pretested and validated and include a copy of the questionnaire as a supplementary information file. Finally, please specify what form of informed consent was obtained by study participants.

Please submit your revised manuscript by Jul 24 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Emily Chenette

Staff Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In the Methods, please describe how the questionnaire was validated. If this did not occur, please provide the rationale for not validating the questionnaire.

- Please include additional information regarding the survey or questionnaire used in the study and ensure that you have provided sufficient details that others could replicate the analyses. For instance, if you developed a questionnaire as part of this study and it is not under a copyright more restrictive than CC-BY, please include a copy, in both the original language and English, as Supporting Information.

- In the ethics statement in the Methods and online submission information, please specify the type of informed consent that was obtained from the participants (for instance, written or verbal, and if verbal, how it was documented and witnessed).

- Please check and revise, as necessary, the p values reported in the manuscript as p cannot equal zero.

3. PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

4.  We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Review notes

Abstract:

The Background section (lines 8-10), should state the problem (which is 2 fold: high burden cervical cancer, low rates of screening) and the objectives of this study, which I also see as two-fold: report prevalence of screening in these districts as well as investigate associations between various socio-demographic predictors and screening.

Methods: line 13 – pick either “univariate” or “bivariate” – they are the same.

Results: Line 15: age – is this relatively older or younger age? Line 15: source of what information? Health information? Line 15/16: the “knowledge” variable needs to be better defined throughout – I would re-define it here as “knowledge of importance of screening to prevent cervical cancer” or as an “understanding of the role of screening in cervical cancer prevention.” Similarly, the “logistics” variable needs to be better defined and I would re-define here as “knowing where to go for screening,” as you do in line 20. The odds ratios (lines 16-21) should be worked in with the prior sentence stating significant predictors. Please add “95%” prior to CI in lines 16-21. Also, please make clear if this is bivariate or multivariate analysis.

Conclusions: The conclusion needs to be re-written. Lines 23-24 only include some of the significant findings. Lines 25-30 go beyond the scope of the data. The first sentence is the overall finding and is good (line 22). I would limit the conclusion to 2-3 sentences in total. For the concluding sentence, I would offer something like: “relatively greater access to resources and knowledge, both disease-related and logistical, predicted a higher prevalence of screening” and then OK to have a final sentence with data-based recommendations for future. The last sentence could be something like: “efforts to increase uptake of screening must address disparities in access to resources and knowledge.”

Background:

In general, the first 2 paragraphs should be combined into 1 (or 2, but re-organized) paragraph(s) painting the picture of cervical cancer in LMICs and Uganda specifically, then you should have a paragraph about what is known about screening, then a description of this particular study’s goals.

Lines 33/34: make 1 sentence about ½ million new cases and 1/3 million deaths annually and cite Bray 2018. Take out the global age-standardized incidence rate as you discuss later (line 50) and it makes more sense there.

Lines 35/37: be careful about saying “disproportionately” as 85% of world’s population lives in LMICs, according to the World Bank, so 70% is not technically “disproportionate.” I think the 70% number is not accurate – The GLOBOCAN estimates more like 84% of new cases and 87% of the deaths occur in LMICs. You cite Torre 2012 as citation #2, but your citation for #1, the Bray GLOBOCAN estimates should replace the 2012 estimates. Take out citation 2 and update.

Line 37: Do you mean “sub”-Saharan Africa?

Lines 38-40: Take out the India statistic and it’s confusing to also hear about breast cancer – I would just focus on the incidence and mortality rates, for which Southern and Eastern African regions are the highest in the world. Be careful: your references #3 and #4 are the same and I’m not sure are complete.

Instead of 41-45: I would end this paragraph by saying something like: “while widespread cytology-based screening in high-income countries has resulted in decreased incidence and mortality from cervical cancer, LMICs, with poor uptake of screening, have not seen similar improvements and, in some cases, incidence and mortality actually continue to rise.” If you want to include the WHO recommendations, I would instead use the WHO guidelines for screening and treatment of precancerous lesions for cervical cancer prevention, 2013. Your statement in line 42 about screening (do you mean once per lifetime screening recommended for 25-49yos?) is an over-simplification of the recommendations. Not sure the SDGs add and reference #6 seems to be incomplete.

Lines 47-64: this paragraph is all Uganda-specific. It’s fine to follow the first general paragraph about high prevalence cervical cancer and low prevalence screening in LMICs, but this paragraph needs to be shorter and more focused:

- Lines 47-49: summarize that Cervical cancer is most common incident and mortal cancer among women in Uganda

- Lines 49-51: one sentence with age-standardized incidence and mortality rates in UG compared to global OK to underscore the magnitude of the problem. Be careful of your citation, I would use the Bruni HPV Uganda-specific report for the citation – make sure the citation is complete (For example, I have cited as: Bruni L B-RL, Albero G, Aldea M, Serrano B, Valencia S, Brotons M, Mena M, Cosano R, Muñoz J, Bosch FX, de Sanjosé S, Castellsagué X. Human Papillomavirus and Related Diseases in Uganda. Summary Report 2016- 02-26: ICO Information Centre on HPV and Cancer (HPV Information Centre), 2016.)

- Lines 52-56 can essentially be cut as long as you have the summary statement about cervical cancer as the most common incident and mortal cancer among women in uganda.

- Line 56-57: cut this sentence about 8/10 women at UCI have cervix cancer. Reference 9: seems incomplete the way it is cited and probably OK to just cite the GLOBOCAN 2018 instead of the Uganda Fact Sheet.

- Lines 57-60: I’m not sure what these lines add to the intro. This is not a paper about cancer, it’s a paper about screening, so I think sufficient to make a grim summary statement about cervical cancer earlier on. I would cut these sentences and references 10-12 (also, these references seem incomplete).

- Lines 60-64: this is a good way to end this paragraph, to bring it back to Uganda. Lines 60-63 should be one sentence about the MOH goal to screen 80% - be careful, your citation for #13 is incomplete: change to “MOH. Strategic Plan for Cervical Cancer Prevention and Control in Uganda, 2010-2014. Kampala, Uganda: Ministry of Health; 2010.” Line 63-64 – what study are you citing here? Citation 14 is incomplete? It might be stronger to start the next paragraph with the best estimate of screening prevalence in Uganda.

Lines 66-85: This paragraph is too long and is disorganized. It is a good idea to use this paragraph to discuss what it is we know about screening and why we need to know more in order to improve screening programs. Some of what is written in this paragraph should be removed and saved for the discussion where you will compare this study’s findings to that of other studies.

- The paragraph could start with your statement about how research on determinants play a role in designing effective interventions (lines 69-70), but its not clear to me that your references 17-21 (careful, most of these references are incomplete and be consistent with either numbering) actually justify this claim – are you suggesting that these studies used some sort of implementation science approach to designing their interventions?

- Lines 72-73 say another general statement about how cervical cancer is preventable if screening is implemented – your citations 22 and 23 don’t really justify this statement which is generally accepted to be true, and the citations are incomplete. This general statement does not belong at this point in the introduction.

- Lines 74-76 and 66-68 go together and given an overview of what is known about cervical cancer screening – be careful of your references, some are incomplete. I would stick to references about Eastern Africa.

- Lines 76-81 focus on urban vs rural – the first sentence is incomplete and unnecessary. The next 2 sentences are too long are not focused (Lines 77-81) – are you just trying to say that living in a rural area is predictive of not screening?

- Lines 81-85 talk about “strengths” of the current study – this goes in discussion, not background. Instead, end the paragraph with a statement about the objective of the current study. Something like, “the objective of the current study was to report the prevalence of screening in Central Uganda and to examine for associations between predictors and successful screening in order to inform the design of future screening programs…”

In general, this is a paper about screening and I am surprised that I don’t find anything in the introduction about VIA versus HPV vs cytology for methods screening. Generally speaking, there is too much about cervical cancer and not enough about screening.

Methods:

The first section should include not only a description of study setting, but also study population – you can include the age-range here… other inclusion/exclusion? Language?

Line 95: reference 34 is incomplete.

Lines 95-97: you claim there is no precedent in the literature, but in line 110, you cite studies which likely report on a prevalence of screening. There are estimates out there – if you think they are underestimates (50% is high), you need to explain why you think prior studies would be underestimate and so you’ve gone with 50%.

Lines 97-98, I think you have overestimated your sample size. You could say something like, “To calculate our sample size, we assumed that if 50% of women in Central Uganda had previously screened for cervical cancer, a total of 402 women would allow us to experimentally determine this proportion within 5% (confidence level 95%) using a binomial “exact” calculation (Hulley SB, Cummings SR, Browner WS, Grady D, Newman TB. Designing clinical research : an epidemiologic approach. 4th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2013. Appendix 6E, page 81). https://www.sample-size.net/sample-size-conf-interval-proportion/

Lines 98-100: this is not a “cluster randomized study,” rather, you used “cluster sampling” to select a simple random sample of clusters. This is very different and needs to be clarified.

Lines 99-101: Why 70% from Wakiso and 30% from Nakasongola? Explain and provide justification.

Line 101: “the target population…” – this belongs in “inclusion criteria.” Reference 35 does not make sense here. If you are justifying the ages in your inclusion criteria, I would cite Ugandan MOH recommendations for screening a particular age range (they recommend 25-49).

Lines 101-103 should be removed here – this is for inclusion criteria. For sampling procedure, describe the cluster sampling technique.

Lines 109-110: Were any of the questionnaires validated? Doesn’t make sense to say that the questionnaire was adapted from “findings” from other studies – sounds like it was modified from questionnaires used in other studies.

Lines 119-124: should be a separate paragraph.

Lines 124-127: the description of the research assistants and their training could be a separate paragraph.

Lines 127-128: talk more about the pre-test/pilot phase – were questions adapted? Community participation is an important part of study design and this could be highlighted here.

Lines 128-129: remove the sentence “the questionnaires were administered by 5…” – this is redundant.

Lines 129-131: this is a good way to summarize, but also, were data electronically captured? Sent to a secure server?

Line 134: a cervical examination is not screening – do you mean VIA? This has to be better defined.

Line 135-136: screening does not test for cervical cancer, but also pre-cancer/dysplasia…

Lines 133-161: this paragraph is way too long and needs to be split up. You don’t need to include so much information about basic demographic variables, just need to say that demographic info including age, religion, place residcen, ethnicity, marital characteristics, parity; as well as health characteristics like use fam planning, recent visit to health center. Lines 147-148: this is the first I’m hearing about this “autonomy” variable – I did not see in abstract, if not used in analysis, don’t need to describe. The description of the “knowledge” variable 149-157 is too long and should be truncated/summarized. Line 157-158 – the justification doesn’t need to be there and the reference 39 is incomplete and doesn’t make sense. Lines 158-161 – description wealth index is too long and also was presented in abstract as dichotomous (rich/poor).

There is description of the “source of information” variable – as that is not an obvious variable, please explain – source of what information, exactly?

Line 164: I don’t understand how Epidata was used, clarify.

Lines 166-171 need to be re-written. I image you used chi-square tests to test for association between outcome (being screened ever) and categorical explatory variables. And I imaging that for continuous explanatory rpedictors, you used t-tests to compare sample means by outcome. I imagine you used logistic regression to explore association demographic and health-related predictors with outcome, adjusting to limit bias from confounding. Discuss how you build the multivariate model and whether sensitivity analyses were performed.

Line 168: you sat that knowledge is a dependent variable, yet you treat it an an independent predictor and your dependent variable is your outcome variable, screening. Fix this.

Results

The results section needs to be rewritten. The section needs to lead with the proportion who had ever been screened (your main outcome).

Tables 1 and 2 should be combined into one table, I would suggest a table that looks like this:

Characteristic Total (%) History prior screening (%) Never been screened (%) OR unadjusted prior screening (95% CI) OR adjusted prior screening (95% CI)

Total (N, %) 845 XXX (21%) XXX (79%)

Age

<=29 315 (37%) 13 87 1.0 1.0

30-39 364 (43%) 23 77 1.4 (0.83-2.19) x.x (a-b)

40-49 166 (20%) 30 70 1.9 (1.05-3.55) x.x (a-b)

Some specific feedback:

Line 192: remove sentence “results in table one…”

Lines 192-200 – the order of the variables needs to be re-organized. For example – Catholics are mentioned in 196 and then protestants are mentioned in line 199. You are describing the population in this paragraph, which is a fine was to start.

Line 202-203 : remove this first sentence. Lines 203- 212: this needs to be made into one paragraph and re-organized. Also, are you discussing bivariate or multivariate analysis? What is going on with the “knowledge” variable? In the Methods section, you discuss a composite score variable, but here there are two knowledge variables – please be consistent.

Lines 223-225 – remove this sentence.

Lines 227-235 – need to be combined and re-written with lines 192-203. This is all the same information.

Some other thoughts about the variables (in table one):

- Age – do you need three categories? Often easier to understand as a dichotomous variable

- Religion – consider combining as Christian versus Muslim

- Study site: note which is urban/which is rural

- Education: consider a dichotomous variable – completed primary (at least some secondary) versus not completed (some primary)

- Occupation: consider professional versus other (“proffessional" is spelled wrong)

- Marital status: consider married vs single/widow/separated

- Ethnicity: consider combining for fewer categories

- Age at first marriage: the “single” category should not be in there.

- Knowledge: this is different than described in Methods and is different table one to table two. Make consistent

- Distance to screening facility: “don’t know where to go” does NOT belong in there – that is a different variable… is that different than “awareness of screening services?”

Discussion

Overall, this is the best-written section of the paper, but it still needs significant work.

Lines 240-246: good to start with your overall finding about screening prevalence. I would take out the first sentence. I don’t think it’s necessarily useful to include stats from Nepal, I would stick to the region.

Lines 248-252: this is about knowing where to go – the logistics variable. I would remove the words “the possible explanation for this finding is that” and just start the sentence with “Increased awareness of …”

Lines 254-259 are about the age variable. This paragraph is OK.

Lines 260-269: this paragraph is about he “source of information” variable. As I’ve mentioned before, this variable needs to be better defined. I would remove the sentence lines 263-265 “this observation might be due to..” I would also remove the sentence lines 267-269 “Secondly it…) I would add a sentence positing that health workers may be essential in health messaging.

Lines 271-274, this paragraph is about the wealth variable – I would re-write the first sentence – take out “corrobortating the results…” part. Overall the paragraph is ok.

Lines 276-281, the “knowledge” variable is problematic and I have brought that up before. I would take out he sentence 278-279 “understanding the impotance…” – this is speculation. Your last sentence is an important point that prior screening is likely related to increased knowledge.

Lines 284-286: Study limitations: can’t assess causality and generalizability. What about sources of bias?

You have two nearly identical paragraphs: recommendations and conclusions – this should just be there once, likely as “conclusions”

Lines 289-295: take out “substantial effect” line 290 and change to something like “associated” – you cannot imply causation. You then focus on the source of information variable and suggest that antenatal providers should discuss screening – this is a really important point and should be developed! Why not screen in pregnancy? What proportion of women interact with providers during pregnancy (at least one anenatal visit)? This also brings in the knowledge variable – be explicit about how these are related. You then focus on the wealth index variable and suggest universal healthcare – what are some challenges there? Does the government have the funding? What would make cervical cancer screening cost effective?

References (lines 317-423) – nearly all the references are incomplete – please be careful and re-do the references.

Reviewer #2: REVIEWER’S REPORT

MANUSCRIPT NUMBER: PONE-D-20-02899

MANUSCRIPT FULL TITLE: Predictors of Cervical cancer screening uptake in Central Uganda

GENERAL COMMENTS

What are the main claims of the paper and how significant are they for the discipline?

Isabirye et al., indicated that they had identified and determined the extent of the association of predictors with screening for cervical cancer among women in two districts of Uganda. These finding provide evidences for the importance of country/community-specific information for the promotion of cervical cancer screening.

Are the claims properly placed in the context of the previous literature? Have the authors treated the literature fairly?

The background is overload with information on cancer distribution (over 60% of the text) while the main focus of the study, which is predictors of screening for cervical cancer are less than 20%. Furthermore, there was nothing about screening, the authors should note that the types of screening available in a community has been show to also be a predictor of participation in screening, particularly in the context of cultural limitation. Again, the problems or concerns associated with the lack of knowledge of the predictors of cervical cancer screening in Uganda have not been discussed at all. Therefore, I suggest the background be written again with much focus on the key words of the objective of the study

Do the data and analyses fully support the claims? If not, what other evidence is required?

The claims are largely supported by the data and analysis, however, the table 1 should be reported in a more informative manner as suggested in the specific comments below. I was wondering why there was no analysis comparing urban to rural women, since the selection of the districts ensure this representation. Was that the difference was not significant?

PLOS ONE encourages authors to publish detailed protocols and algorithms as supporting information online. Do any particular methods used in the manuscript warrant such treatment?

No method used in this study warrants a publication of detailed protocol as supporting information online

If a protocol is already provided, for example for a randomized controlled trial, are there any important deviations from it? If so, have the authors explained adequately why the deviations occurred?

None of such protocols was identified in this manuscript

If the paper is considered unsuitable for publication in its present form, does the study itself show sufficient potential that the authors should be encouraged to resubmit a revised version?

Yes, the authors are encouraged to make the suggested revisions

Are original data deposited in appropriate repositories and accession/version numbers provided for genes, proteins, mutants, diseases, etc.?

The Manuscript gives the indication that all data are fully available without restriction upon reasonable request from the corresponding author.

Are details of the methodology sufficient to allow the experiments to be reproduced?

Details of the method for the selection of the 2 of the 27 districts, how the number of villages selected from each district was arrived at, and the selection of women within each village need to be provided.

Is the manuscript well organized and written clearly enough to be accessible to non-specialists?

The manuscript is well organized and clear

SPECIFIC COMMENTS

LINE 1: The title should read as “Predictors of cervical cancer screening uptake in two districts of Central Uganda”

This because the 2 districts have not been shown to be representative of the 27 district from which they were selected, therefore, the data should not be generalized to Central Uganda. It does not have the claimed external validity.

LINE 13: It will be more appropriate to rather state the type of statistic used to analyse what variable or for what purpose or which outcome.

Lines 29-30: No data/finding has been provided to support this conclusion, this should not be a conclusion of this study.

LINE 37: You mean "Sub-Saharan Africa region"

Lines 41-45: These statement are hanging. Authors need to the question "and so what?" in respect to this statement.

Lines 51: “25” should be “25.0”, and “Global” should be “global”

Lines 81-82: This should state clearly the overall aim of the study, not just a part of it.

Line 83: how was this minimisation achieved?

Lines 90-92: How were the two districts selected from the 27 districts? How many of the 27 districts were rural and urban/peri-urban? Provide more information regarding this selection for the reader to be convinces the selection was representative and fair.

Lines 98-99: the study was earlier on described as a cross-sectional study, so why call it a cluster random study here? Furthermore, there is no description of how the two districts were selected, and no grouping (clusters) have been described.

Lines 99-101: what informed this proportions?

Line 102: Please recheck this fact and determine if it applies to all countries?

Lines 104-105: Why 24 and 10 villages? How were the women recruited from the villages?

Line 128: Were these "nearby community" near by the study area or the place the study was designed? Please state the names and how far they are from the study districts or villages. This is to convince the reader that there was a very low probability of contamination of the study villages.

Line 128-130: Is this per village or per day?

If each of the 5 RAs collected data from 6-8 participants in each village and 34 villages were involved in this study then 1020 - 1360 participants were involved. However, if each of the 5 RAs collected data from 6-8 participants per day and 28 day were used for data collection then 840 - 1120 participants were involved, which accounts for only 850 participants in the study?

Line 178: Was the voluntary informed consent written or oral?

Line 193: was this proportion not designed to be so as indicated in the methods? If not, that is this was observed after recruitment, then it should be taken out of the methods. By the way it was reported as 70% in the method.

Lines 192-200: some of the numbers do not have the “%”

Line 218 (Table 1): Table should be improved. I suggest a proper cross-table as indicated below. All p value stated as “0.000” should be stated as “<0.0001”

Screening, n (%) Chi-squared p-value

Ever screened Never screened Subtotal (%)

Age group 0.001

< 29 n1 (%) n2 (%) n1+ n2(%)

30-39

40-49

Total

Line 225: Were all variables used in the multivariable logistic analysis? How were they entered into the multivariable analysis?

Lines 244-246: These seem inappropriate, comparing a region to a hospital what is the purpose of the comparison? Were the studies similar to this study, if not indicate the differences.

Lines 255-256: Rather, state what the other similar studies found, what this study found is already stated in the results.

Line 268: Please provide a reference.

Lines 289-290: I suggest this is deleted, it is not a recommendation.

Lines 293-295: there is no finding in this study that relates to this recommendation. who are the recommended being directed at?

Lines 297-298: No need to repeat the results in the conclusion, Delete “Only 1 in 5 women (21%) had ever screened for cervical cancer.”

Lines 300-305: how different is this from a recommendation, as stated earlier? Please delete.

Line 350: check reference, seems incomplete

Line 351: check numbers in author name

Line 365: check et al

Line 398: upper case

Line 413: check reference, seems incomplete

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Megan Swanson

Reviewer #2: Yes: Adolf Kofi Awua

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: swanson_reviewnotes_isabirye.docx

Click here for additional data file.

16 Aug 2020

We appreciate your critical reflection on our scientific piece. Thank you.

Submitted filename: Response letterPLOSOne.docx

Click here for additional data file.

11 Nov 2020

PONE-D-20-02899R1

Predictors of cervical cancer screening uptake in two districts of Central Uganda

PLOS ONE

Dear Dr. Isabirye,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

Please respond to additional comments by a reviewer.

==============================

Please submit your revised manuscript by Dec 26 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Clement A. Adebamowo, BM, ChB Hons; FWACS, FACS, ScD, FASCO

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Reviewer Notes

Predictors of cervical cancer screening uptake in two districts of Central Uganda

Overall, this is a much stronger version. Thank you for working to revise the manuscript. The following are minor suggestions for revision.

Abstract:

- I think you can remove lines 13-14

- Line 16, after “targeted interventions”, I would add “to improve screening”

- Line 26, as you are shifting to conclusion, I would add the word “Overall,” cervical cancer screening low…

- Lines 29 and 30: remove “the above findings suggest that” and start this sentence with “Efforts to increase uptake…”

Background: In general, this is much improved with the emphasis on screening, rather than cancer, but I still think the section could be shortened.

- I would take out some of the cancer-specific stats from paragraph one and transition quicker to the problem of inadequate screening. Consider splitting into two paragraphs.

- Line 57: need to write-out “Visual inspection with acetic acid” before VIA

- Line 64: You already defined LMICs

- Lines 63-67: consider removing these lines and starting with “the success of screening depends on uptake…” – also depends on follow-through the cascade to treatment. Vaccination is beyond the scope of your discussion. If you are going to mention vaccination, I would say something to the effect of “while comprehensive vaccination is cost-effective and lifesaving, incidence of cervical cancer not expected to drop for at least 2 decades after widespread vaccination uptake (Jit M, Brisson M, Portnoy A, Hutubessy R. Cost-effectiveness of female human papillomavirus vaccination in 179 countries: a PRIME modelling study. The Lancet Global health. 2014) and in the meantime cervical dysplasia and early cancers will remain common and treatable”

- Lines 72-79: Be careful, comparing various screening methods is beyond the scope of this paper – the debate between HPV testing and VIA is nuanced and there are strong feelings on both sides.

Methods:

- Line 181-182: in forward-stepwise model building, usually cast a wider net and include predictors with p value more like 0.2 or less and also those known to be associated a priori. Just a comment.

- Lines 183-184, shorten to: “multicollinearity tests were performed”

Results:

- Lines 201 – 207 – you can just refer to table for characterization of the demographics.

- I had previously suggested combining tables 1 and 2. OK to have separate tables, but I would recommend using bivariate ORs, rather than chi-square tests.

- Lines 226-232 – careful – it’s not 1.93x more likely, but, rather, higher odds – this error is repeated.

- Table 2: add 1.0 for the OR for the referent groups. Personally, I don’t think you need p-values if you are giving CIs.

Discussion:

- line 239: you said “1 in 5” before and now saying “2 in 10”

- line 255: type “messaging”

Conclusion:

- lines 297-299: careful about the antenatal screening recommendation - ASCO specifically recommends NOT screening in pregnancy (I disagree with the recommendation and write an opinion piece about it which was published in JGO) – I would take a step back here and suggest that one approach to increasing the provider-pt health education could be to utilize times when reproductive age women are already interfacing with healthcare, like pregnancy.

- Be careful talking about health insurance coverage in Uganda – tie the sentiment to your findings, that “rich” women were more likely to get screened – thus, your recommendation is for screening opportunities to be expanded specifically for poor women – If you are going to talk about national financing for the health sector, this is a huge topic and I would likely rather just hint at it here.

- Seems like your main findings were that rich women, women who learn from providers and women with high knowledge were more likely to screen. But seems like the conclusion only really touches on the first 2 – do you want to somehow also bring in the knowledge variable here?

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Megan L Swanson

Reviewer #2: Yes: Dr Adolf Kofi Awua

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Submitted filename: swanson_reviewnotes1025_isabirye.docx

Click here for additional data file.

17 Nov 2020

Response letter

Date 15 November 2020

To: PLOS ONE

From: "Alone Isabirye" aloneisab@gmail.com

Subject: Response to review comments of our manuscript submitted to PLOS ONE (PONE-D-20-02899R1)

PONE-D-20-02899R1

Predictors of Cervical cancer screening uptake in two districts of Central Uganda

Alone Isabirye, Martin Kayitale Mbonye, Betty Kwagala

PLOS ONE

Dear Editor,

Thank you for your reply regarding our manuscript "Predictors of Cervical cancer screening uptake in two districts of Central Uganda" (PONE-D-20-02899R1). We are grateful for the reviewers’ comments. We have revised and modified the manuscripts according to the referees’ critiques. As a consequence, we provide a revised manuscript with the reviewers’ suggestions integrated therein:

Response to editor’s comments

-Deposit laboratory protocols in protocols.io; Not applicable

-Uploading our figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool; Not applicable

Response to reviewer 1’s comments

The overall comment about the revised manuscript is appreciated.

Abstract:

- Lines 13-14 were removed.

- Line 16, we added “to improve screening” after “targeted interventions”

- Line 26, as we are shifting to conclusion, we added the word “Overall,” cervical cancer screening low…

- Lines 29 and 30: we removed “the above findings suggest that” and started the sentence with “Efforts to increase uptake…”

Background:

- We removed some cancer-specific statistics from paragraph one especially about Uganda and the Cancer Institute and transitioned quicker to the problem of inadequate screening. Additionally, the paragraph was split into two paragraphs.

- Line 57: we wrote-out “Visual inspection with acetic acid” before VIA

- Line 64: second definition of “LMICs” was removed

- Lines 63-67: The section was revised while considering the proposed recommendations by the reviewer.

- Lines 72-79: Phrase comparing various screening methods was removed.

Methods:

- Line 181-182: The comment of casting a wide net in variable consideration is appreciated. We included some variables known to be associated a priori like type of residence. However, almost all variables left out in our multivariate model had p-values far higher than 0.2.

- Lines 183-184, the statement was shortened to: “multicollinearity tests were performed”

Results:

- Lines 201 – 207 – The table was referred to.

- We are pleased with your decision to accept the two tables. We also humbly propose that chi-square test is used at bi-variate level. This is because data was analyzed at three levels and the selection of statistical approaches at the respective levels was guided by Bloom’s taxonomy of learning objectives.

- Lines 226-232 – 1.93x more likely was corrected to “higher odds” – this error was fixed throughout the manuscript.

- Table 2: 1.0 was added for the OR for the referent groups. The authors preferred presenting the results with both p-values and confidence intervals because these statistical measures complement one another.

Discussion:

- Line 239: stating “1 in 5” before and stating “2 in 10” later was systematized by mentioning “1 in 5” throughout.

- line 255: We correctly typed “messaging”

Conclusion:

- Lines 297-299: The antenatal screening recommendation was fixed as suggested by the reviewer.

- Our recommendation regarding poor women was fixed as proposed by the reviewer.

- We included high knowledge among our main findings.

Response to reviewer 2’s comments

- No comments.

We hope that our modifications render our manuscript in its current form suitable for publication in PLOSONE

Yours sincerely,

Isabirye Alone

aloneisab@gmail.com

On behalf of the authors

Submitted filename: .Response to Reviewers.docx

Click here for additional data file.

19 Nov 2020

Predictors of cervical cancer screening uptake in two districts of Central Uganda

PONE-D-20-02899R2

Dear Dr. Isabirye,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Clement A. Adebamowo, BM, ChB Hons; FWACS, FACS, ScD, FASCO

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

23 Nov 2020

PONE-D-20-02899R2

Predictors of cervical cancer screening uptake in two districts of Central Uganda

Dear Dr. Isabirye:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Clement A. Adebamowo

Academic Editor

PLOS ONE