**4. Discussion**

Analyzing data from the two COVID-19 hotspot regions in Cameroon revealed a significant drop in DTP-1 coverage of 3.3% and in DTP-3 coverage of 5.4%. This drop resulted into 8247 and 12,896 children missing out on their DTP-1 and DPT-3 vaccines, respectively, in the pandemic period compared to the pre-pandemic period. Moreover, the DTP series vaccination dropout rate increased from 7.5% to 9.3% in the Centre Region and from 3.7 to 6.0 in the Littoral Region. Our findings run contrary to national administrative data, which suggested improved childhood vaccination access and utilization during the pandemic period. Improvement at the national level is understandable, because vaccination has mainly been driven by the organization of periodic intensification of routine immunization (PIRIs) in many districts of some regions, particularly those in the Southwest and Northwest regions [17]. Moreover, several vaccination campaigns were organized in other regions in response to VPD epidemics, coupled with the introduction of a second dose of measle and rubella vaccine [18,19]. This may have had a significant impact on the coverage of other antigens, increasing national vaccination coverage.

Despite this improvement in national coverage, which was purely driven by the PIRIs, our findings clearly align with those reported by the third round of the global pulse survey on the impact of COVID-19 on immunization services [6]. The survey revealed that 70% (64/91) of participating countries reported disruptions in routine immunization services, with 18% (16/91) experiencing severe disruptions between February and August 2020 [20]. Moreover, a study conducted in a tertiary hospital in Cameroon revealed a decreased demand for childhood immunization services during the COVID-19 pandemic, with a significant drop in the coverage of DTP-containing vaccines [16]. Other studies reported a similar decline in immunization indicators during the pandemic [21,22].

The significant drop in vaccination coverage in our study can be explained by the advent of a novel pandemic that encountered an unprepared and weak health system, hence the grave challenges in meeting the demands of pandemic control. This led to task shifting in favor of the pandemic, creating an unintended negative impact on essential health services, such as routine immunization. Additionally, sub-optimal training of clinicians regarding routine patient care, including vaccination amid the pandemic, created, the fear of contracting COVID-19 when offering health services [9]. This fear was worsened by inadequate personal protective equipment and standards of operation to keep the disease in check in a clinical setting [9]. Delays in COVID-19 confirmatory diagnosis due to limited test kits and diagnostics targeted every patient presenting in a clinical setting with upper or lower respiratory tract symptoms as a suspected case, leading to poor care, even for patients presenting with other ailments [9]. This complexity and uncertainty associated with contracting the COVID-19 infection may have created a spillover effect of COVID-19 stigma and hesitancy toward other routine essential health care services, including immunization services.

Although districts recorded varying degrees of change in vaccination access and the utilization of tracer indicators between the pre-pandemic and pandemic period, more than two-thirds of them reported a drop in DTP-1 (62.5%) and DTP-3 coverages (71.4%). Up to a 31.6% and a 43.7% drop in DTP-1 and DTP-3 coverage, respectively, was reported in some districts. This finding lends support to the results of a cross-sectional study in Senegal that showed a significant decrease in immunization uptake at the health facility level [23]. This further emphasizes the need for a real-time assessment tool to be used at the different tiers of vaccination service delivery, including health facilities. This is important because aggregated data at higher administrative levels may mask prevailing low performance at lower operational levels. The role of such a tool in data-driven decision making at all levels is invaluable.

The region most heavily hit by the COVID-19 pandemic (the Center Region) recorded a higher drop in vaccination access and utilization in the pandemic period, with more districts reporting a drop in vaccination indicators compared to the results from the Littoral Region. The drop ranged from 5.7% (*p* < 0.0001) to 0.8% (*p* = 0.0002) in the Centre and Littoral regions, respectively. There was also a significant drop in the utilization of immunization services in both regions, and estimates stood at 7.6% (*p* < 0.0001) and 3.1% (*p* = 0.0003) in the Center and Littoral regions, respectively. In a country such as Cameroon, with limited health resources, this piece of information may be helpful to prioritize regions and districts with higher decline in RI indicators, as this may guide the development of a post-COVID-19 recovery plan to reverse the impact of the pandemic on key RI indicators. This finding can also be employed in informing policy on future pandemic management.

Despite the potential usefulness and application of our findings, there are certain limitations that must be acknowledged. These limitations are essentially linked to data completeness on the DHIS-2 platform, which was the main source of data for our study. Based on the DHIS-2 data quality assessment tool, data completeness in the pre-pandemic period was 96% and 100% in the Littoral and Centre regions, respectively; however, during the pandemic period, data completeness was at 91% and 94% in the Littoral and Centre regions, respectively. In this study, we adjusted this limitation by weighting the data against regional data completeness. The data weighting may have introduced bias in some districts by disproportionately increasing or decreasing vaccination coverage.

Based on our findings, we will first recommend a further survey in a sample of these districts to identify factors associated with the decline in vaccination coverage during the pandemic. Second, we recommend the development and validation of a digital tool that can support the early detection of the impact of a pandemic on RI variables at all health system tiers. These two recommendations may be valuable in developing tailored strategies to detect and reverse-inverse trends of the pandemic on RI performance.
