Epidemiology of human papillomavirus among women living with HIV in rural Uganda: a retrospective cohort study Authors: Kabali Bwogi, Justine Kimbowa Nakiwala, Ivan Kasamba, Christopher Bwanika, Joseph Baruch Baluku, Jane Nakawesi, Catherine Senyimba, Ronald Mulebeke, Barbara Mukasa Published: 2026 ABSTRACT AbstractBackgroundPersistent human papillomavirus (HPV) infection is a significant precursor to cervical abnormalities and invasive cancer. While the causal relationship between specific HPV types and cervical cancer is established, little is known about the distribution of HPV types among women living with HIV (WLWH) in Uganda. We aimed to describe the epidemiology of HPV types among women living with HIV at Mityana and Kiboga Hospitals in rural central Uganda.MethodsA retrospective review of demographic and clinical data, including age and antiretroviral therapy (ART) history, was conducted among WLWH tested for HPV between July 2021 and July 2022. GeneXpert HPV DNA real-time Polymerase Chain Reaction assay (RT-PCR) detected HPV 16, 18/45, and other pooled high-risk types (31, 33, 35, 39, 51, 52, 56, 58, 59, 66, and 68). HPV infections were categorized into single and multiple subtypes. Statistical analysis involved Chi-square tests and modified Poisson regression with robust standard error models to evaluate associations and identify factors linked with HPV infection.ResultsAmong 1305 WLWH, the overall prevalence of HPV infection was 39.1%. HPV16 and HPV18/45 were the most common genotypes, with a prevalence of 8.1% and 8.2%, respectively. Multiple HPV infections were observed in 18.4% of cases, with combinations of HPV 16, HPV 18/45, and other high-risk HPV types. Increasing age was significantly associated with a lower prevalence of HPV infection (PR = 0.97; 95% CI: 0.95–0.99; p = 0.003).ConclusionA high prevalence of genital high-risk HPV infection was observed among WLWH attending ART clinics in public health facilities in Uganda. Other high-risk HPV types were the most frequently detected, compared to HPV18/45 and HPV16. The high prevalence of high-risk HPV among WLWH, particularly among younger age groups, highlights the need to integrate HPV screening within routine HIV care services and to strengthen targeted prevention strategies, including vaccination and early screening in rural settings.Keywords: Human papillomavirus types, High risk, Women living with HIV, Prevalence, Oncogenes Background Globally, women living with HIV (WLWH) have a significantly increased risk of cervical cancer [1, 2]. In 2018, the risk of cervical cancer among WLWH was six times higher than among women in the general population [3]. Global estimates show that 5.8% of new cervical cancer cases in 2018 were among WLWH, and 4.9% were attributable to HIV infection, with Eastern and Southern Africa (SEA) being the most affected regions [4]. High cervical cancer prevalence among WLWH has been observed in southern (63.8%) and eastern (27.4%) Africa [5, 6]. Cervical cancer is understood to mainly be caused by persistent human papillomavirus (HPV) infection [2, 7, 8]. HPV exists in more than 120 different genotypes; however, only 13–15 are found in cervical cancers and other cancers and are called ‘high risk’ HPV (HR-HPV) [9–11]. HPV 16 and 18 are associated with two-thirds of all cervical cancers as well as subtypes of cancer of the vulva, vagina, anus, oropharynx, and skin [9, 12]. Globally, HPV 16 is linked to approximately 50% of cervical cancers [13]. HIV enhances HPV-induced carcinogenesis [6, 14], increasing both the risk and the early development of cancer [2, 15]. Compared to HIV-negative women, cervical cancer mortality is two times higher among WLWH [2]. Prior studies in Uganda have reported a high prevalence of HR-HPV among both HIV-negative and WLWH. One study reported prevalence ranges of 10.2–40% and 37–100% among HIV-negative women and WLWH, respectively [9]. Cervical cancer is the most diagnosed cancer among women in Uganda, with WLWH at particularly high risk. Despite recommendations by the Ministry of Health Uganda for annual screening since 2010, access and uptake have remained low, with limited integration of screening services into HIV care and frequent late referrals for treatment. In response, the Ministry of Health Uganda initiated a national scale-up of cervical cancer screening in 2020, aligned with the World Health Organization strategy to eliminate cervical cancer by 2030 [16]. However, limited data exist on the distribution of high-risk HPV genotypes among women living with HIV in rural Uganda, highlighting the need for context-specific evidence to inform targeted prevention strategies. Materials and methods Study design and setting We conducted a retrospective cohort study, reviewing programmatic records from WLWH who underwent HPV testing at two public district hospitals in central Uganda (Mityana and Kiboga District Hospitals). Mityana Hospital, also Mityana General Hospital or Mityana District Hospital, is a hospital in the town of Mityana, Mityana District in the Central Region of Uganda, the hospital is located in the central business district of the town of Mityana, about 85 km (53 mi) east of Mubende Regional Referral Hospital, the regional referral hospital [17] and Kiboga General Hospital, also Kiboga Hospital, Kiboga Main Hospital, or Kiboga District Hospital, is a hospital in the Central Region of Uganda. The hospital is located on the Kampala–Hoima Road, in the central business district of the town of Kiboga, about 79 km (49 mi) southeast of Hoima Regional Referral Hospital, in the city of Hoima and serves parts of the following districts: Kiboga, Kyankwanzi, Kibaale, Nakaseke, Mubende, and Hoima [18]. Each hospital serves a catchment population of approximately 100,000 people and is accredited to provide comprehensive HIV care services. In addition, they offer a wide range of healthcare services, including inpatient care with designated wards for men, women, and children, an emergency surgery operating theatre, and blood transfusion services [19]. Study population We collected all available records of WLWH ≥ 18 years, enrolled in the Antiretroviral Therapy (ART) care program, had an HPV test between July 2021 and July 2022, and test results reported as either “positive” or “negative” for HPV 16, 18/45, and other HR-HPV (31, 33, 35, 39, 51, 52, 56, 58, 59, 66, and 68). HPV testing was performed using the GeneXpert HPV DNA Assay, which is part of routine HIV care for WLWH at these hospitals. Women younger than 18 years were not eligible for HPV testing under the prevailing testing guidelines and were consequently excluded from this study, despite being part of the population of women living with HIV. The GeneXpert HPV DNA assay used in this study detects only HR-HPV genotypes, including HPV16, HPV18/45, and other pooled high-risk HPV types. Therefore, all HPV-related outcomes reported in this study refer specifically to high-risk HPV infections. Test results were extracted from electronic medical records and laboratory health management information systems, which are integrated into the hospitals’ digital healthcare infrastructure. This electronic system ensures accurate and efficient data collection, storage, and retrieval. Study variables We abstracted available secondary data from electronic medical records and laboratory health management information systems, including HPV test results and genotypes, clinical data: ART initiation and current regimen, viral load, body mass index (BMI), WHO clinical staging, and demographics: age and duration on ART. The outcome variable was high-risk HPV infection (positive vs. negative). Independent variables included age, ART duration, viral load status, CD4 count, BMI, ART regimen, and WHO clinical stage. Statistical analysis The primary outcome was HPV infection, defined as having a record of any positive HPV test result. Frequency distributions and proportions were utilized to describe demographic characteristics and HPV genotypes among WLWH. We evaluated factors associated with HPV infection using a modified Poisson regression with robust standard errors, since the outcome prevalence exceeded 10% [20]. Modified Poisson regression with robust standard errors was used to estimate prevalence ratios, with HPV infection as the dependent variable and selected demographic and clinical characteristics as independent variables. Factors were incorporated into the model based on hierarchical relationships, beginning with distal factors and progressing to proximate ones. Crude and adjusted prevalence ratios (PR) and 95% confidence intervals (CI) were calculated. Statistical significance was determined at the 5% level. Analyses were conducted using Stata version 12.0 (StataCorp, College Station, TX). Results Social demographic characteristics of the WLWH in Uganda A total of 1305 records from WLWH were included in the analysis. The median age was 36 years [interquartile range (IQR) 30–42]. Over half of WLWH (n = 911, 69.81%) had been on antiretroviral therapy (ART) for over five years. Table 1 shows characteristics of study participants. Table 1Clinical characteristics of women living with HIV at two public hospitals in UgandaVariableCategoryMedian (Q1, Q3)Total n (%)Body mass index24.13(21.2,27.4)Art category*(n = 1304)DTG based1289 (98.85)Non DTG based15 (1.15)Viral load status*(n = 1272)Non-Suppressed37(2.91)Suppressed1235 (97.09)HIV WHO clinical stage*(n = 1182)Stage 1864 (73.10)Stage 2255 (21.57)Stage 353 (4.48)Stage 410 (0.85)CD4 category* (n = 699)< 200124 (17.9)> 201574(82.1)*Missing data, n-number of participants, %percentage, Q1-1st quartile, Q3rd quartile Clinical characteristics of women living with HIV at two public hospitals in Uganda *Missing data, n-number of participants, %percentage, Q1-1st quartile, Q3rd quartile The participants’ median body mass index (BMI) was 24.13 (IQR: 21.22–27.43). Most participants, 1289 (98.85%), used a Dolutegravir (DTG)–based regimen. Additionally, 1235 (97.09%) WLWH had a suppressed viral load, and 864 (73.10%) were classified under HIV WHO clinical stage 1. Prevalence of HPV among women living with HIV in Uganda The prevalence of HPV among 1305 WLWH was 39.1% (n = 510, 95% CI; 36.46–41.76). Among the HPV-positive patients (n = 510), 211 came from Kiboga Hospital, with the remaining 299 from Mityana Hospital. Pooled HR-HPV subtypes were detected in 399 cases (30.57%, 95% CI, 28.13–33.13). The prevalence of specific oncogenic sub-types HPV16 was 8.12% (n = 106, 95%CI; 06.76–09.73) and HPV18/45 at 8.20%, (n = 107, 95%CI; 06.83–09.82) respectively. HPV distribution among different categories of women living with HIV in Uganda The HPV distribution among different participant categories revealed that most HPV patients (n = 113, 22.26%) were aged 30 to 34. The proportion of participants with HPV was significantly different in all age categories, with a P-value = 0.001. The highest prevalence for HPV16 was among participants aged 35–39 (n = 25, 23.6%), while HPV18/45 was most common among those aged 40–44 (n = 26, 24.0%). Of all age groups, 399 individuals between the ages of 30 and 34 were found to have other HR-HPV (n = 96, 24.1%). Most HPV patients had been on ART for over five years (60.4%). There was a significant association between ART duration and testing positive for HPV (P < 0.001). The WLWH, who had been on ART for more than five years, had the highest number of cases within various subgroups of HPV. Within this group, 71 individuals (67.0%) had HPV16, 63 individuals (58.9%) had HPV18/45 and 234 individuals (58.7%) had other HR- HPV. Across all BMI categories, most participants with HPV had a BMI of 24.9, 212 (47.6%), though there was no significant association between BMI category and having HPV (p-value: 0.790). Also, the BMI category of 18.6 to 24.9 had the most significant number of participants with HPV subgroups, 46 (50.0% for HPV16, 44 (47.3% for HPV18/45), and 167 (48.1%) for HR-HPV. 501 (98.2%) of the 510 participants with HPV were on a DTG-based ART regimen, but there was no significant association between the ART category and HPV (P-value: 0.240). Also, the DTG-based regimen contained most participants with HPV across all HPV subgroups, but this had no significant association. Distribution characteristics of single and multiple infections The analysis of HPV infections in all the samples revealed that among the 510 individuals, 20.39% (104) were infected with a single type of HPV, while 18.43% (94) had multiple HPV infections. Among those with multiple HPV infections, 60.94% (78 out of 128) had a combination of HPV 18/45 along with other high-risk HPV types, 28.13% (36 out of 128) had a combination of HPV 16 along with other high-risk HPV types, and 10.94% (14 out of 128) had a combination of HPV 16 and 18/45 (Table 2). Table 2Distribution of HPV among different categories of women living with HIV in UgandaVariableHPV n (%)HPV16 n (%)HPV18/45 n (%)HR-HPV n (%) Age Group (years) 15–194 (0.8%)1 (0.9%)1 (0.9%)3 (0.8%)20–2438 (7.5%)8 (7.6%)5 (4.7%)32 (8.0%)25–2999 (19.4%)16 (15.1%)22 (20.6%)81 (20.3%)30–34113 (22.2%)24 (22.6%)19 (17.8%)96 (24.1%)35–3997 (19.0%)25 (23.6%)24 (22.4%)68 (17.0%)40–4488 (17.2%)18 (17.0%)26 (24.0%)64 (16.0%)45–4948 (9.4%)8 (7.6%)7 (6.5%)35 (8.8%)50 and above23 (4.5%)6 (5.7%)3 (2.8%)20 (5.0%) Duration on ART < 0.9 months13 (2.5%)0 (0.0%)4 (3.7%)11 (2.8%)1–4.9 years189 (37.1%)35 (33.0%)40 (37.4%)154 (38.6%)≥ 5 years308 (60.4%)71 (67.0%)63 (58.9%)234 (58.7%) BMI Category < 18.531 (7.0%)7 (7.6%)9 (9.7%)21 (6.1%)18.5–24.9212 (47.6%)46 (50.0%)44 (47.3%)167 (48.1%)25–29.9136 (30.6%)30 (32.6%)28 (30.1%)105 (30.3%)>=3066 (14.8%)9 (9.8%)12 (12.9%)54 (15.6%) ART Category DTG-based501 (98.2%)103 (97.2%)105 (98.1%)392 (98.3%)NON-DTG based9 (1.8%)3 (2.8%)2 (1.9%)7 (1.8%) Distribution of HPV among different categories of women living with HIV in Uganda Further investigation into multiple infections revealed that the age group between 35 and 39 had the highest infection rates for both HPV 16 + 18/45, the age group between 30 and 34 had the highest infection rates for HPV + other high-risk HPV types, and the age group between 40 and 44 had the highest infection rates for HPV 18/45 + other high-risk HPV types (Fig. 1). Shows the distribution of HPV types Factors associated with HPV among women living with HIV in Uganda Factors associated with HPV among WLWH are presented in Table 3. In bivariate analysis, increasing age was significantly associated with a lower prevalence of HPV infection (PR = 0.97; 95% CI: 0.95–0.99; p = 0.003). Participants presenting with WHO clinical stage III or IV had a significantly higher prevalence of HPV infection compared to those in stage I or II (PR = 1.73; 95% CI: 1.01–2.96; p = 0.044). Other factors such as ART duration, ART regimen category, CD4 count, viral load status, and BMI category showed no statistically significant associations with HPV prevalence in bivariate analysis (p > 0.05 for all). In multivariate analysis, age remained independently associated with lower HPV prevalence (adjusted PR = 0.97; 95% CI: 0.96–0.99; p < 0.001). Although WHO clinical stage III/IV had a higher prevalence ratio compared to stage I/II (adjusted PR = 1.58; 95% CI: 0.95–2.61), the association was not statistically significant (p = 0.077). Table 3Multivariate analysis of factors associated with HPV among 1305 WLWH in UgandaVariableCategoryPrevalence ratio (PR) (95% CI)p-valueMultivariate PR (95% CI)p-valueAgeContinuous0.97 (0.95–0.99)0.0030.97 (0.96–0.99)0.000ART Duration> 10 Years< 4.9 Years0.96 (0.61–1.53)0.871––5–9.9 years0.75 (0.50–1.12)0.166––ART StatusDTG – BasedNon-DTG – Based1.24 (0.61–2.52)0.550––CD4 Category> 201< 2000.88 (0.57–1.35)0.558––Viral LoadSuppressedNon-Suppressed1.40 (0.80–2.44)0.235––BMI Category18.6–24.925-29.90.81 (0.58–1.11)0.191––30 above0.93 (0.63–1.39)0.730––< 18.51.02 (0.55–1.89)0.949––WHO StageStage I & IIStage III & IV1.73 (1.01–2.96)0.0441.58 (0.95–2.61)0.077 Multivariate analysis of factors associated with HPV among 1305 WLWH in Uganda Discussion This retrospective cross-sectional analysis of all eligible records from two rural public hospitals found a high HR-HPV prevalence among WLWH attending ART clinics. Pooled high-risk HPV types accounted for the largest proportion of detected infections, exceeding HPV 18/45 and HPV 16, and increasing age was independently associated with a lower prevalence of HPV infection. Because the analysis used routine facility data and a single recorded HPV result, the findings should be interpreted as prevalence estimates rather than evidence of persistence, clearance, or causation [21]. Our findings revealed a high prevalence of HPV infection, with 39.1% of women testing positive. This prevalence rate was higher compared to previous studies conducted in Uganda [22–24] and other African countries [25, 26], which ranged between 17% and 25%. However, studies focusing on WLWH showed prevalence rates in line with the present study, ranging from 30% to 64% [25–29]. Taken together, these data suggest that the burden identified in the present study is neither anomalous nor trivial; rather, it sits within the upper-middle range of recent estimates for WLWH in sub-Saharan Africa. Direct numerical comparison should nevertheless be cautious because studies differ in urban versus rural setting, age structure, underlying HIV control, and laboratory platforms, including whether genotypes are individually identified or pooled into non-16/18 channels [21]. The predominance of pooled other HR-HPV types is one of the most important findings. It mirrors the recent Ugandan national study, in which other hrHPV types constituted 68.8% of infections, and is broadly consistent with recent evidence showing substantial contributions from HPV58, HPV52, HPV35, HPV45, and other non-16/18 types [20]. This regional pattern matters because it indicates that service demand in WLWH is not driven solely by HPV16 and HPV18. At the same time, the present study used a GeneXpert assay that reports HPV16 and HPV18/45 individually while pooling the remaining high-risk types into a single channel. Accordingly, the data demonstrate that non-16/18 types contribute materially to the screening burden, but they cannot identify which individual genotypes are responsible for that pooled signal. The age pattern is also meaningful. In the present study, the highest absolute burden of infection was observed among women aged 30–34 years, and the study showed a modest but statistically significant decline in HR-HPV prevalence with increasing age [9, 11, 30]. The most careful interpretation is that younger WLWH in these settings carry a greater detectable infection burden. However, individual HPV genotypes 16 and 18 peaked in the 35–39 and 40–44 age groups, respectively. This may reflect a combination of earlier acquisition, slower clearance in some women, and differences in exposure patterns across the life course [26, 30, 11]. This underscores that age is an important determinant of HPV risk among WLWH and suggests that prevention and screening efforts in rural Uganda should particularly target younger WLWH to reduce the HPV burden. The high prevalence of HPV among WLWH is of great concern due to the increased risk of developing cervical cancer. Persistent infection with high-risk HPV types is a key factor in developing cervical intraepithelial neoplasia and cervical cancer. The most common genotypes identified were the pooled high-risk HPV [11, 26, 31]. Therefore, future development of multivalent preventive HPV vaccines should consider including additional high-risk HPV genotypes beyond the currently targeted types 16/18 [9]. Uganda, through the Ministry of Health, has led efforts to combat the threat of HPV by making vaccinations available; however, this initiative has largely been unsuccessful, resulting in low vaccination uptake [32, 33]. Therefore, approaches that involve men in the fight against the spread of HPV infections would be warranted, as studies in Uganda have shown a high incidence of 66.5 per 100 person-years in HIV-positive men and 32.9 per 100 person-years among HIV-negative men, with a prevalence of 90.7% among men living with HIV and 60.9% among HIV-negative men [34]. The study highlighted an association between a longer duration of ART and a higher prevalence of HPV infection among women who had been on ART for five years or longer, and showed a higher prevalence of HPV. This finding is consistent with previous studies reporting an increased risk of HPV persistence and progression to cervical dysplasia among women on long-term ART [23, 35]. Despite the positive impact of ART on immune function and longevity in WLWH, cervical cancer remains high [36]. WLWH experience increased prevalence and incidence of cervical HPV infections and higher risks of persistent HPV infections, subsequent cervical intraepithelial lesions, and cervical cancer [37]. Furthermore, WLWH are more likely to have high-risk HPV infections characterized by higher prevalence rates, co-infection, persistence, and reactivation [38, 39]. It is important to note that the association between ART duration and HPV infection may be multifactorial, involving both immunological and potential behavioral factors such as sexual behavior and adherence to ART. This line of research holds valuable insights that could potentially aid in developing efficacious biological therapeutics targeting HPV and HPV-related diseases. Moreover, such advancements may have significant implications for preventing HPV-related cancers in individuals infected with high-risk HPV strains. Strengths of the study include the comprehensive nature of data collection, which provided valuable insights into the prevalence, distribution, and associated factors of HPV infection among women living with HIV in Uganda. The inclusion of clinical characteristics such as body mass index, ART category, viral load status, and WHO clinical stage enriched the understanding of the cohort and allowed for more robust interpretations. Several limitations should be acknowledged in the study. Firstly, the cross-sectional design limits the ability to establish a causal relationship between the identified factors and HPV infection. Longitudinal studies are necessary to understand the dynamic nature of HPV infection in this population. Secondly, the study was conducted in specific healthcare settings in Uganda, which may restrict the generalizability of the findings to other populations or regions. Additionally, the use of pooled high-risk HPV results limited the ability to perform genotype-specific analyses beyond HPV16 and HPV18/45, thereby restricting a more detailed understanding of the distribution of individual high-risk HPV types. The reliance on routinely collected programmatic data may have introduced misclassification, incomplete records, or missing data for some variables, which could affect the accuracy of the findings. Further research in diverse settings must validate these findings and ensure their applicability to broader populations. In conclusion, High-risk HPV prevalence was high among women living with HIV in these rural Ugandan hospitals, with other pooled high-risk HPV types accounting for the largest proportion of infections. Younger age was associated with a higher prevalence of infection, with the greatest burden observed among women aged 30–34 years. These findings support integrating HPV screening into routine HIV care and strengthening targeted cervical cancer prevention programmes, particularly for younger women living with HIV, alongside continued scale-up of national screening services. Acknowledgements We are grateful to all the patients whose data were used in the study. We thank Dr. Richard Muhindo for the insightful advice during the writing and review of the manuscript. Abbreviations Antiretroviral Therapy Body Mass Index Centers for Disease Control and Prevention Confidence Interval Dolutegravir Eastern and Southern Africa Electronic Medical Record High-Risk Human Papillomavirus Human Immunodeficiency Virus Human Papillomavirus Sub-Saharan Africa Women Living with HIV World Health Organization Author contributions BK conceived the study. BK, JKN, IK, CB, JBB, JN, CS, RM, and BM participated in the study design and data collection. BK and IK participated in data analysis. All authors participated in writing and reviewing the manuscript. All authors read and approved the final manuscript. Funding This research received support from the US President’s Emergency Plan for AIDS Relief (PEPFAR) via the Centers for Disease Control and Prevention (CDC), grant number 1NU2GGH002046-01–00. The views expressed in this manuscript are those of the author(s) and do not necessarily reflect the official stance of the funding organizations. Data availability The clinical data sets used and /or analyzed during this study are available from the corresponding author upon request. Declarations Ethics approval and consent to participate The study was approved by the Mildmay Uganda Research and Ethics Committee (REC REF 0804–2018), as well as the U.S Centers for Disease Control and Prevention (U.S CDC) under 45 C.F.R. part 46.101(c); 21 C.F.R. part 56. A waiver of consent was granted to use secondary data because the data were sourced from a publicly accessible Electronic Medical Records (EMR) platform. All methods were performed in accordance with the Declaration of Helsinki. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement The clinical data sets used and /or analyzed during this study are available from the corresponding author upon request.