Cervical Cancer Draft 3-8
Cervical Cancer Draft 3-8
Cervical Cancer Draft 3-8
Burden : In 2018, there were 570,000 new cases of cervical cancer and 311,000 deaths from cervical cancer
worldwide with 50,000 deaths occurring in Sub-Saharan Africa (Arbyn, GBD Data Tool). Short-term
consequences of cervical cancer include medical expenses and treatment side effects. Long-term side effects
include lower quality of life, lowered family stability, decreased productivity, and death (Ashing-Giwa). Cervical
cancer is the leading cause of cancer-related deaths in Botswana due to limited screening and early detection,
and accounts for 26% of cancers in women (Mingo, Grover). A 2015 study in Botswana found that most
women have little knowledge of risk factors and when to seek screening. In a population of women already
attending clinics in Botswana, 72% never had any cervical cancer screening (Mingo). Women in Botswana are
at a particularly high risk for developing cervical cancer due to 3.2% of them having HPV. (HPV center). In
high-income countries, education and organized screening have led to low cervical cancer incidence. In
contrast, the lack of education and organized screening in Botswana has led to a high disease burden in which
166 women in Botswana die each year from cervical cancer (Grover). There is limited data quantifying the
influence of education on uptake of cervical cancer screening (Grover). Therefore, we propose a clinic-based
randomized trial in Botswana to determine whether cervical cancer education will increase cervical cancer
screening rates and improve knowledge of cervical cancer among women aged 18-60 years. Our objectives
are to:
Aim 1: Determine if a clinic-based cervical cancer education intervention increases cervical cancer screening
uptake among women aged 18-60 years attending public health clinics. Hypothesis 1: Clinics randomized to a
cervical cancer education intervention will have higher uptake of cervical cancer screening compared to clinics
not receiving the education intervention over a period of 1 year.
Aim 2: Demonstrate that a clinic-based cervical cancer education intervention increases knowledge about
cervical cancer among women aged 18-60 attending public health clinics. Hypothesis 2: Women attending
receiving cervical cancer education interventions will have increased knowledge about cervical cancer risk
factors and preventative measures after the educational presentation.
Study Overview: Cervical cancer has a high burden in Botswana and an educational intervention equipping
women with knowledge of screening could increase early screening and detection of cervical cancer. This
study will implement a CC education intervention in selected clinics in Botswana. We will compare CC
screening rates at clinics receiving the education intervention to clinics waitlisted for receiving the intervention.
Study Design and Sites: This randomized controlled trial will evaluate the effectiveness of clinic-based
education to increase knowledge and uptake of cervical cancer screening to improve early identification of CC.
The trial will study cervical cancer rates at 16 public health clinics in mid-sized cities across Botswana located
in Mmadinare, Palapye, Bobonong, Sefhare, Francistown, Gantsi, Gweta, Hukuntsi, Masunga, Thamaga,
Gumare, Kasane, Rakops, Tutume, Tsabong, and Letlhakane. To establish a baseline screening rate estimate
for each site, the first month of the trial will be a “run-in period” and will not include the education intervention,
but will collect data on the number of eligible women attending clinics and the rate of women aged 18-60
getting cervical cancer screenings. The trial will compare screening uptake rates between control clinics
Eligibility: Clinics who consent to providing de-identified aggregate data to determine the site-specific number
of eligible women and number of cervical cancer screenings, and to implementing the educational intervention
are eligible for this study. Women aged 18 - 60 years attending the clinic who have cervixes (no record of a
hysterectomy) are considered eligible. Women who have received the cervical cancer education intervention
previously will be ineligible to receive it again, and women who do not receive the entire education presentation
will be excluded.
Randomization: Each clinic and clinic catchment area will be cluster randomized to education (ED) (N=8) or
education intervention waitlist (W) (N=8). Randomization will be generated by a biostatistician using random
1:1 block size.
Blinding: Study healthcare workers and target population will not be blinded, but those analyzing outcome
data will be blinded to cluster allocation.
Enrollment: Clinics will record the number of women who check in at the clinic who fit the eligibility criteria,
regardless of the reason for attending clinic. The pre-education and post-education questionnaires will be given
to eligible women who give oral consent to having their responses recorded as part of the study.
Intervention: Researchers will train a cohort of health workers at each clinic to give CC educational
presentations. Health workers will be given presentation scripts on cervical cancer education in English and
Tswana. The presentations will be given in a designated education presentation room in the clinic once in the
morning and once in the afternoon everyday in English, immediately followed by the same presentation in
Tswana. Before and after the presentation, health care workers will administer and collect questionnaires. In
addition, every eligible woman will be given an educational pamphlet (in English or Tswana) on cervical cancer
when she checks in for an appointment. Pamphlets will also be available in the waiting area of intervention
clinics for patients to take.
Control Arm: The control arm will consist of only the standard of care (SOC), which is the Pap smear cervical
cancer test, which is available at every clinic. SOC does not include any education intervention. Clinics
included in this arm will be placed on a waitlist to receive the intervention after the trial is completed, after the
one month run-in period and the year-long study.
Outcome Assessment: The primary outcome of cervical cancer screening uptake will be measured with a
ratio of aggregate de-identified data of the number of eligible women that got screened out of number of
eligible women that visited the clinic in the year following the 1 month “run-in period” in both the control and
intervention arms. This data will be recorded directly by the clinic and abstracted by the study team. We are
measuring this outcome because if there is an increased rate of screening, there is an increased chance of
detecting cervical cancer early. If there is a statistically significant difference between the control and
intervention groups’ ratios, it will be attributable to the education intervention. The secondary outcome is
knowledge of cervical cancer among eligible women attending the clinic. This outcome will be measured using
questionnaires about cervical cancer screening facts given to eligible women before and after educational
presentations. Improved knowledge about cervical cancer will be defined as a higher average post-
presentation questionnaire score compared to the average pre-presentation score.
Limitations & Anticipated Problems: The first limitation is that the study population is restricted to women
already coming to the clinic. Therefore our results are not generalizable to the rest of the female population
that do not attend clinics regularly or within the timespan of the intervention. This limits the generalizability
across socioeconomic groups, and limits the generalizability to women in urban regions. Secondly, There is
also a chance for spillover in two different ways. One way that spillover affects exposure is that those in the
intervention group may be in contact with women attending SOC clinics, and may share some of the things
they learned from the intervention to the standard of care group. This would cause an increase in screening
uptake at SOC clinics. Another source of spillover is if women who receive the intervention at one of the
designated clinics gets screened at another clinic, which will be an issue of misclassification of the outcome,
and will also result in increased screening uptake at SOC clinics. In addition, because the intervention will
happen at certain times of the day, not everyone who visits the clinic will observe the presentation, and only a
fraction of the patients at the intervention clinic will actually receive the intervention. To help mitigate this effect,
all eligible women will still receive educational pamphlets. If the correlation between education and screening
uptake is strong enough, a difference between SOC and the intervention group will still be significant despite
the aforementioned limitations.