Knowledge, attitude, and practice of cervical cancer screening among women living with HIV in the Kilimanjaro region, northern Tanzania

Abstract Background Cervical cancer is the fourth most common cancer globally among women in incidence and mortality. Women living with HIV (WLHIV) are disproportionately at a higher risk of developing the disease. Aim To determine the knowledge, attitude, and practice of cervical cancer screening among WLHIV in the Kilimanjaro region, northern Tanzania, following the integration of these services in routine HIV care in the country. Methods and results A cross‐sectional study was conducted in the Kilimanjaro region among 297 WLHIV attending care and treatment centers (CTC) in Hai district and Mawenzi regional hospitals in northern Tanzania between 21 August and 3 September 2020. A questionnaire was used for data collection using face‐to‐face interviews. Data were analyzed using SPSS version 20.0. Frequencies and percentages summarized categorical variables and numerical variables summarized using median and interquartile range (IQR). About half (50.2%) of 297 WLHIV in this study had ever screened for cervical cancer, and 64% screened within the past 12 months preceding the survey. Although 90% ever heard of cervical cancer screening, only 20.5% knew when WLHIV should start screening. Over half (52.5%) had adequate knowledge of prevention, 38.4% on risk factors, and 27.9% of cervical cancer signs and symptoms. Two‐thirds (66.7%) had positive attitudes toward cervical cancer screening. A major source of cervical cancer screening information was the health care providers (80.1%) and the mass media (66%), particularly radio. Conclusions The WLHIV in this study had inadequate knowledge but favorable attitudes toward cervical cancer screening, while half had screened for cervical cancer. Efforts should be directed to capacity building of health care providers at CTC and scaling up the mass media campaigns as relevant interventions to promote the uptake of cervical cancer screening programs among WLHIV in Tanzania.


| INTRODUCTION
Cervical cancer is a public health concern being the fourth most common cancer among women in incidence and mortality, 1,2 with 570 000 new cases in 2018 representing 6.6% of all female cancers worldwide. 2, 3 Approximately 90% of all new cervical cancer cases occur in low-and middle-income countries and is also the second most common cancer among women in sub-Saharan Africa. 1,2 In 2018, 26 009 women were living with cervical cancer in East Africa, which accounted for 26.2% of all cancer cases among women in the region, Tanzania accounting for 39% of all cases. 1 The prevalence of cervical cancer among women living with HIV is high and more alarming in sub-Saharan Africa. Prevalence is estimated to range between 1.3% in Kenya and 6% in Nigeria. [4][5][6][7][8] Within Tanzania, the prevalence ranges from 7.3% in Mwanza to 11% in Morogoro, 9,10 higher than that reported in Nigeria. Poor screening practices may partly contribute to the high prevalence of cervical cancer among WLHIV in Tanzania.
Virtually, all cervical cancer cases (99%) are associated with genital infection with high-risk human papillomavirus (HPV)-a widespread virus transmitted through sexual contact. 2,11-13 WLHIV have a higher risk of developing the disease, mainly due to their immunecompromised state. 1,9,10,[14][15][16] Knowledge and attitudes toward cervical cancer screening are crucial in determining the screening intervention's uptake among women. [17][18][19][20] The belief of not being susceptible to cervical cancer, fear of cancer diagnosis, fear of exposing their (women's) private parts, anticipated pain of the testing procedure, a long distance from home to the clinic, poor access to screening results, long waiting time, and fewer healthcare workers are other barriers to cervical cancer screening among women living with HIV. 18,[21][22][23] Despite the heavy burden of the disease, cervical cancer is a highly preventable disease in women, including those living with HIV. 19,24 WHO introduced comprehensive cervical cancer prevention in 2014, which comprises cervical cancer screening, targeting women who are at higher risk of developing the disease. 24 With the introduction of mass HPV vaccination for young girls in some developing countries, there are opportunities to offer the vaccine to HIV-positive middle-aged women through the existing HIV care and treatment programs. 16 In Tanzania, cervical cancer screening has been integrated into HIV care and treatment services where screening is initiated soon after HIV diagnosis without regarding the woman's age and is conducted annually. 25 Despite that, the proportion of HIV-positive women screened for cervical cancer in CTC is low. In Dar Es Salaam, only 9% of women living with HIV ever had at least one cervical cancer screening test. 26 However, this study was conducted before the integration of cervical cancer screening in routine HIV care. Since this integration, little has been done to assess the uptake of cervical cancer screening services among women living with HIV in Tanzania. This study aimed to determine the knowledge, attitudes, and cervical cancer screening practices among WLHIV in the Kilimanjaro region, northern Tanzania, following the integration of these services in CTC services. Findings from this study will provide information to assess the efficacy of this program and inform future interventions. These findings may also assist the healthcare providers, particularly at CTC, in promoting the uptake of cervical cancer screening services among WLHIV. Moreover, the study results may contribute to developing policies, guidelines, and strategic decisions that will enhance the current screening practices in this population.

| Study design, setting, and population
We conducted a health facility-based cross-sectional study in the Kilimanjaro region between 21 August and 3 September 2020. The region has 396 health facilities that are 20 hospitals, 41 health centers, and 335 dispensaries. Out of these health facilities, 52 provide care and treatment (CTC) services. 27,28 CTC is the gateway where people living with HIV can access HIV care, treatment, and support services. 25 The prevalence of HIV in the Kilimanjaro region was 2.6% and was high (3.1%) among women aged 15 years and above. 29,30 The study population was all WLHIV in the Kilimanjaro region and were attending CTC at data collection time. The study included all women aged 18 to 55 years who provided informed consent. In Tanzania, WLHIV are supposed to be screened for cervical cancer immediately after HIV diagnosis. 25 The study excluded severely ill women and had undergone total hysterectomy because severely ill women could not respond to the questions. The women with total hysterectomy had their cervix surgically removed.

| Sample size and sampling
The sample size was calculated using the formula for estimating a single proportion, given as (N = [Z a/2 ] 2 × p[1−p]/e 2 ), where N is the desired sample size, and p is the estimated prevalence of cervical cancer screening among women living with HIV, assumed to be 20%. Furthermore, e is the margin of error or precision (5%), and Z is the standard normal value (1.96) corresponding to a 95% confidence interval. After adding a 10% proportion of nonresponse, the minimum estimated sample size was 271 participants.
A simple random sampling technique was used to select Hai among the rural districts of the Kilimanjaro region. Moshi municipality was purposefully selected to ensure rural-urban representativeness.
One CTC in each district with the highest number of women enrolled was selected (Hai district hospital and Mawenzi regional referral hospital). All women who attended the CTCs were selected for inclusion.
Sampling was done proportional to the size of each selected CTC.

| Data collection methods
Face-to-face interviews were used for data collection using an electronic administered questionnaire. The questionnaire was adapted and modified from previous studies. 31,32 The questionnaire was in both English and Swahili languages. It contained information on participant social-demographic characteristics, knowledge, and attitudes on cervical cancer screening, cervical cancer screening practices, and HIV care and treatment. Trained doctor of medicine students collected data. The interviews were administered in Swahili (local) language and were conducted in a quiet place around the CTC clinics after obtaining informed consent. Each interview took about 20 to 30 minutes.

| Study variables
The primary outcome was cervical cancer screening practice measured by asking women if they had ever screened for cervical cancer or not, the reason for screening, the timing of screening since diagnosed with HIV, and whether they had ever screened in the past 12 months.
Knowledge and attitudes on cervical cancer screening were secondary outcomes. Knowledge of cervical cancer was measured by asking participants if they ever heard cervical cancer and knowledge of causes, signs, risk factors, and prevention. Knowledge of causes was measured using five-item questions, the signs using 11 items, risk factors using 12 items, and prevention using 5 items. Each of these items carried one point when answered correctly and zero points when wrongly answered. Final scores were categorized into good knowledge (≥50% of the scores) and poor if otherwise. 13 Attitude on cervical cancer screening was measured by asking the participant 10 questions concerning thoughts and feelings toward cervical cancer screening. 32 The mean score was used to categorize the respondents into positive vs negative attitudes. 33 The independent variables included social demographic characteristics and information on HIV. Social-demographic variables included age in years, date of birth, number of children, the highest level of education (no education, primary, secondary, and higher education), marital status (single/ never married, married/cohabiting, divorced/separated/widowed, specify if others), occupation

| Data analysis
Data cleaning and analysis were performed using SPSS version 20.0.
Frequencies and percentages were used to summarize categorical variables and means/medians and standard deviations/interquartile range for numeric variables. The findings were summarized into tables, graphs, and narrations.

| Participant background characteristics
A total of 303 women living with HIV met the inclusion criteria and were invited to participate in the study. Only 297 of those invited consented to participate in making a response rate of 98%. The median age of 297 women living with HIV who participated in this study was 44 (IQR 36-49.5) years. About half (49.2%) of the women were 45 years of age or above. More than half (58.2%) were widowed/divorced or separated, and 60.6% were self-employed. About 89.2% of the women reported having no health insurance, and only 5.1% reported a history of cervical cancer in their families (Table 1).

| Cervical cancer screening practices
Half of all women living with HIV in this study reported having ever screened for cervical cancer. Of these, 64% were screened within the Moreover, almost half (52.5%) had adequate knowledge of preventing cervical cancer (Table 3).
Of the 297 women living with HIV in this study, 80.1% received information on cervical cancer screening from the health care providers and only 8.1% from print media (8.1%) (Figure 1).

| Attitude toward cervical cancer screening
The majority (88.9%) of the women living with HIV in this study saw the need for cervical cancer screening, and over three-quarters (78.5%) were willing to be screened without having signs and symptoms. About two-thirds (66.7%) of the women had a positive attitude toward cervical cancer screening, and 71.4% were comfortable to be screened by any health care provider regardless of their gender (  43 This could have been among the reasons for the high screening uptake. The lower screening uptake in Ethiopia despite high proportions of adequate knowledge may also be due to inadequate health care provider cervical counselling on cancer screening. Health care workers and mass media (radio and television) were the primary sources of information concerning cervical cancer screening, similar to Dar es salaam. 26 However, this was so to only a fifth (20%) in Nigeria. 33 A higher percentage of participants who received information from health care workers in Tanzania can also be due to the integration of cervical cancer screening programs and CTC service delivery. In Nigeria, cervical cancer information has not been incorporated into the HIV test and counselling services. Other sources of information in Nigeria included family members, relatives, and friends.
This calls for strengthening the health care workers in CTCs and enhancing mass media education campaigns across the country to increase awareness and uptake of cervical cancer screening.
About two-thirds of women living with HIV in this study had a positive attitude toward cervical cancer screening, which is significantly higher than 20.8% reported in Dar Es Salaam. 26 The difference may be because Koneru et al 26 conducted the study before integrating cervical cancer screening with HIV services at CTC in 2017. 25 Furthermore, this study's estimate is higher than 43.5% reported in Nigeria but less than 87% in Ethiopia. 33

ACKNOWLEDGEMENT
We acknowledge the regional medical officer of the Kilimanjaro Region for permission to carry out this study. We also appreciate the support from the CTC health care providers in Mawenzi regional hospital and Hai district hospital during the data collection process. We dearly appreciate the study participants whose consent and participation enabled the availability of data used in this study.

CONFLICT OF INTEREST
The authors have stated explicitly that there are no conflicts of interest in connection with this article.

AUTHOR CONTRIBUTIONS
All authors had full access to the data in the study and take responsi-

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.

ETHICAL STATEMENT
The study was approved by the Kilimanjaro Christian Medical University College Research and Ethics Review Committee (KCMU-CRERC) and received an approval number UG 090/2020. Oral informed consent was obtained from participants before participation in the study.
Participation in this study was voluntary and did not affect the patient's routine CTC cervices. Participants were allowed to refuse to answer any questions and terminate the interview when they desired. Confidentiality of information was ensured by using unique identifiers.