Study design
This is a cross-sectional descriptive study that was carried out among HIV-positive women who were attending the adult HIV clinic in Jos University Teaching Hospital (JUTH), Jos between November2018 and January 2020 .
Study Area
JUTH is a tertiary health institution located at Jos, Plateau state with a catchment area of the four states of the north-central geopolitical region of Nigeria. These are Benue, Nassarawa, Kaduna and Bauchi states. Plateau state is located in Nigeria's middle belt with an area of 26,899 square kilometers and an estimated population of about three million people in 2019. [20]
Study Population
The study population comprised HIV-positive women attending the HIV clinic in the hospital. All HIV-positive women attending the clinic were included while those with a history of cervical cancer, previously treated premalignant lesions of the cervix and those that were pregnant were excluded from the study.
Data Collection
De-identified data of HIV-positive women attending the adult HIV clinic in JUTH for the first time was obtained from the secured electronic records of the clinic. The cervical Pap smear cytology screening outcomes were reported according to the Bethesda 2001 cytology reporting system. [21]
The key dependent variable was the cervical cytology (Pap smear) screening outcomes. Data on sociodemographic and reproductive characteristics as well as clinical information such as nadir viral load and nadir CD4 counts were also obtained. The age, CD4 count and viral load were used as continuous variables while categorical variables were parity (0, 1–4, ≥ 5), age (≤ 43,>43), age at first sexual debut (< 15, 15–18, ≥ 19), number of sexual partners (1, 2–4, ≥ 5) and use of contraception (used, not used).Viral load was further log transformed for analysis in view of the skewed nature of the measurement. Pap smear results was categorized into normal squamous intraepithelial lesion (SIL) and abnormal SIL while the abnormal SIL was further categorized into mild dysplasia and severe dysplasia. Abnormal SIL were ASCUS, ASC-H, LSIL, HSIL, and HSIL with suspicion of invasion and AGUS while mild dysplasia and severe dysplasia were ASCUS, AGUS, LSIL and ASC-H, HSIL, HSIL with suspicion of invasion respectively.
Data analysis
The data collected was analyzed with STATA software, version 14 college station (Stata Corporation, TX, USA; 1985). Continuous variables were expressed as means, medians, standard deviations and interquartile range and categorical variables as percentages and proportions. Logistic regression models were developed to explore the association of selected variables with cervical cytological outcomes among the women.
Three level of analysis were done. The first level analysis was a descriptive analysis to determine the overall prevalence and pattern of cervical cytological abnormalities based on various sociodemographic and clinical characteristics of the women. Differences in proportions and means of observable measures between the cervical cytology outcomes were assessed using Chi-square test for proportions and t-test for difference of means.
The second level analysis was bivariate logistic regression analysis performed to assess the association between apriorically selected sociodemographic, clinical (viral load and CD4 count), reproductive characteristics and (1) cervical cytology results (normal and abnormal squamous intraepithelial lesion) (2) mild dysplasia (3) severe dysplasia. The third level analysis was a multivariate logistic regression analysis involving all the factors that were significantly associated with cervical cytology (Pap smear) results at the second level analysis. Box plots graphs of ages of these women by Pap smear results was plotted. P < 0.05 was considered as statistically significant. The results are presented as crude odds ratios (CORs) or adjusted odds ratios (AORs) and their 95% CI.
Methodology
We performed a cross sectional study of adult women with HIV who presented for Pap smear screening test at the Jos University Teaching Hospital, Jos, North Central Nigeria between November 2018 and January 2020. Prevalence of abnormal cervical finding was reported in percentage and also expressed per 100 population of women. Association between characteristics of the women and finding of pap smear screening was determined using logistic regression with 95% confidence interval and crude odd ratio used as interval and point estimates of the effect of the characteristics of the women on the outcome of the screening while a p -value of < 0.05 was considered statistical significant.