Context. Cervical cancer (CC) is a leading cause of death among women living with HIV (WLHIV) in resource-limited settings. Yet, effective methods for screening and preventing CC are available. The recommanded approach for CC screening is based on multiple steps, including initial test to detect human papillomavirus (HPV) infection, visual inspection to identify women with HPV at risk for precancerous lesion and treatment when required. Dropout may occur at these different steps, compromising the success of the CC elimination strategy. Performing all the screening and treatment sequences in a single visit has been recommanded based on the results of a large South African trial. Yet, in many contexts, including those with limited resources, the screening and treatment activities are performed in multiple visites for logistical reasons, resulting in many dropouts. Different strategies for delivering screening with HPV testing for WLHIV are possible. A first approach (centralized approach) consists of having well equipped reference centres with experienced health workers and referring women to these centers. An alternative consists of having a mobile unit who can bring equipment and health workers and perform the CC screening in the usual places of patient care (decentralised or mobile team approach). Each of these two approaches has advantages and limitations in terms of coverage, completeness, cost and quality of screening. It is necessary to evaluate them in real life to inform national decision-makers on the best strategy to use in their countries. The OptiTri-MU study aims to evaluate and compare the effectiveness of these two strategies for delivering CC screening (centralized screening versus decentralized screening). It will also assess the implementation of each strategy and include three sub-studies designed to evaluate : * the performance of urinary HPV testing. * the performance of different methods to identify women requiring a treatment. * the risk of post-treatment cervical disease. Design This is a trial in which the intervention (mobile team) will be implemented gradually. All sites starts with the centralized screening strategy. At each period, a new site is ransomly selected and start the the decentralized screening strategy. There will be 6 periods of 10 weeks. The effectiveness of the intervention will be assessed by comparing the outcomes at each site before and after implementation and by comparing the sites with each other. The primary outcome for effectiveness is the screening completeness 120 days after enrollment. The study will also assess the implementation of each screening strategy in terms of : * Success through the measure of fidelity, reach and completeness * Identification of adaptation, barriers and facilitating/leverage factors * Perception, feasibility and acceptability of the screening strategies (by patients and health care workers) Other study objectives include : * To assess the performance of different methods to identify women requiring a treatment * To assess the performance of HPV testing on a urine sample compared with vaginal self-collected or cervical (clinician-collected) samples * To assess the efficacy of treatment in terms of post-treatment cervical lesions Study population. Participants will be WLHIV aged 25 to 49 and eligible for CC screening. Health care workers will also be invited to participate to the implementation research. The data collected will be quantitative and qualitative.