An Evaluation of the Adverse Events Following Voluntary Medical Male Circumcision Surveillance System in Zvimba District, Zimbabwe, 2020

Background: The Adverse Events following Voluntary Medical Male Circumcision Surveillance System (VMMC AESS) serves to identify adverse events (AEs) for safe provision of VMMC. In 2019, Zvimba District reported two adverse events out of 542 circumcisions translating to a 0.4% AE rate against an acceptable threshold of 2% and this is suggestive of underreporting. We evaluated the system to verify existence of underreporting of AEs, assess the surveillance system attributes and evaluate data quality. Methods: A descriptive cross-sectional study was conducted using the updated Center for Disease Control (CDC) guidelines for surveillance system evaluation. We proportionately sampled VMMC client intake records and purposively selected health workers involved in the VMMC programme into the study. An interviewer-administered questionnaire was used to collect data from 14 30 December 2020 on knowledge of the VMMC AESS among health care workers and surveillance system attributes. We used a checklist to check for data quality. Data were analysed using Epi Info 7 statistical software. Results: Of the 31 health workers that participated in the study, 21 (68%) had fair knowledge of the surveillance system. Out of 384 records reviewed, 104 had missing data on AE monitoring days 7 and 42. Sixteen (52%) of participants regarded filling AE reporting forms as difficult and the process as time-consuming. The surveillance system was not representative as only two out of eight facilities were conducting outreaches in hard-to-reach areas. Conclusion: There was possiHow to cite this paper: Hlabangana, T., Mugurungi, O., Govha, E., Juru, T.P., Gombe, N.T., Chadambuka, A., Omondi, M. and Tshimanga, M. (2022) An Evaluation of the Adverse Events Following Voluntary Medical Male Circumcision Surveillance System in Zvimba District, Zimbabwe, 2020. World Journal of AIDS, 12, 1-19. https://doi.org/10.4236/wja.2022.121001 Received: January 13, 2022 Accepted: February 27, 2022 Published: March 2, 2022 Copyright © 2022 by author(s) and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/


Introduction
Voluntary Medical Male Circumcision (VMMC) is defined as the voluntary removal of the prepuce (foreskin) of the penis for medical reasons [1]. Male circumcision (MC) reduces the risk of female-to-male HIV transmission by up to 60% and since 2007, the World Health Organization (WHO) and the US President's Emergency Plan for AIDS Relief (PEPFAR) began supporting the implementation of voluntary medical male circumcision(VMMC) programmes across 14 priority African countries including Zimbabwe [2] [3].
Zimbabwe embraced voluntary medical male circumcision in 2009 as part of a comprehensive package of HIV prevention and treatment services [4]. The adverse events following voluntary medical male circumcision surveillance system in Zimbabwe were adopted at the inception of the programme in the country in 2009 to ensure safe and quality VMMC service provision. It is a paper-based recording and reporting system and the flow of information in the surveillance system is shown (Figure 1). The flow of information in the surveillance system as depicted in Figure 1, begins with case-based identification of adverse events at the health facility where four copies of the VMMC adverse event reporting form are completed by the health care workers involved in VMMC (for moderate and severe AEs). One copy of the form is kept at the health facility at which the AE is identified with the rest of the forms being forwarded to the district office, the provincial medical directorate's office and subsequently to the head office. Information about reportable AEs from the health care facilities is reported to the district office within 24 hours and a summary of monthly AEs is reported to the technical partner within 30 days. Severe A/Es are supposed to be reported to the national office within 48 hours. Whilst there are studies that have been done before in Zimbabwe in evaluating the adverse events in VMMC surveillance system, these studies have been specific to the PrePex device [2] [3]. There has been a gap in evaluating the effectiveness of the surveillance system in the VMMC programme as a whole when other circumcision methods which are currently in use in the country such as the dorsal slit and surgical resection methods are utilised [5].  There were only two adverse events reported from 542 circumcisions conducted, accounting for 0.4% against an acceptable threshold of 2% for adverse event rates. These statistics are suggestive of underreporting of adverse events in the district. We evaluated the surveillance system to verify the existence of underreporting of adverse events and to ensure that the surveillance system attributes are functioning effectively so that no adverse events are missed should they occur.

Study Design
We conducted a descriptive cross-sectional study in Zvimba District from using the updated Centre for Disease Control (CDC) guidelines for surveillance system evaluation and collected data from 14-30 December 2020 [6]. The minimum sample size of 250 records for review was calculated based on assumptions that completeness of filling forms was 80% (expected national target for completeness), precision of the study at 20%, Confidence level 95% based on a study by Dube et al. (2017) in Chegutu District, Zimbabwe [9].

Sampling and Study Participants
The primary respondents were health care workers directly involved in voluntary medical male circumcision namely registered general nurses (RGNs), primary care nurses (PCNs) and government medical officers (GMOs). We recruited all the eight health facilities currently offering VMMC in the district into the study. We purposively selected health care workers directly involved in the circumcision programme who were on duty during data collection and proportionate random sampling of VMMC client intake records were done at each health facility.
Data Collection and Analysis A pre-tested interviewer-administered questionnaire (see Appendix 1) was used to collect data from health workers on VMMC surveillance system attributes namely stability, simplicity, flexibility, acceptability and representativeness. We also collected data on knowledge on the surveillance system from health care workers. The questionnaire was pretested and adjusted by interviewing five health care workers involved in VMMC (10% of the sample size) at Norton Hospital which is about 39 km from the capital city Harare. We chose Norton Hospital as it has a similar setup with the health care facilities offering VMMC in Zvimba District. Adjustments that were made included adding a section that assessed health care workers' knowledge on classification of adverse events according to type, severity and timing. We reviewed VMMC client intake forms covering the period from 1 January 2019 to 31 December 2020 at each health facility to check for data quality and to check if there were any unreported adverse events.

Measurement of Variables
Knowledge of health care workers We assessed health workers' awareness and understanding of the components of the AEs following VMMC surveillance system by asking participants to answer questions regarding VMMC AEs surveillance. The questions centred on ability to state the objectives for AEs surveillance, classification of the AEs, the reporting channel, the time frame to report and the criteria for coding of adverse events. A three-point Likert scale was used to grade knowledge level among health care workers.

Usefulness
This is a measure of whether the system contributes to the prevention and control of adverse health-related events, including an improved understanding of the public health implications of such events [6]. Usefulness was assessed by ascertaining whether there were any meetings done to review AE data and whether there were any public health actions that were implemented based on the surveillance system findings.

Data Quality
Data quality refers to the completeness and accuracy of the data recorded in the surveillance system. We measured data quality by checking the completion of data on various fields on the form and the capturing of data on monitoring AEs on days 2, 7 and 42 of the review periods.

Simplicity
This attribute refers to the structure and the ease of operation of the surveillance system [6]. We assessed simplicity by looking at the proportions of health workers who had ever identified an adverse event and completed adverse event reporting forms. We also evaluated simplicity by assessing whether the health care workers felt they needed further training on filling AE data and how they regarded the simplicity of filling out AE reporting forms.

Acceptability
Acceptability refers to the willingness of persons and organizations to participate in the surveillance system [6]. We assessed acceptability by ascertaining willingness to continue participating in the surveillance system by health workers among other indicators. We assessed acceptability objectively by checking for the availability of VMMC monthly return forms that were submitted to the next level. Stability refers to the reliability (i.e. the ability to collect, manage, and provide data properly without failure) and availability of resources needed for the surveillance process [6]. We assessed stability by looking at the availability of resources required to run the surveillance system and the adequacy of knowledge of health care workers on adverse events data reporting. Objectively, we used a checklist to assess the availability of resources such as functional vehicles, stationery and communication tools at the facilities.

Representativeness
This measures whether the surveillance accurately describes the occurrence of a health-related event over time and its distribution in the population by place and person [6]. We assessed representativeness by looking at the availability of VMMC outreaches in hard-to-reach areas and whether there were any private health facilities offering VMMC. pandemic, the essential infection prevention protocols were strictly adhered to.

Demographic characteristics of VMMC health care workers in Zvimba
District, 2020 Thirty-one health care workers were recruited into the study with females constituting 16 (52%) and males 15 (48%) of the study participants. The majority 18 (58%) of study participants were Registered General Nurses (RGNs) and = 42). The median time in service for participants was 12 years (Q 1 = 10, Q 3 = 15) and the median time in-service involved with VMMC was 4 years (Q 1 = 2, Q 3 = 5) ( Table 1).
Knowledge of VMMC health care workers on the surveillance system All the participants were aware of the VMMC AEs surveillance system and could state at least one objective of the surveillance system. However, on classification of severity and coding of AEs, less than half of participants 13 (42%) gave the correct description on both criteria. Overall, using a three-point Likert scale, the majority 21 (68%) of the health care workers had fair knowledge whilst 6 (19%) had good knowledge and only 4 (13%) had poor knowledge of the surveillance system ( Table 2).
Stability of the surveillance system in Zvimba District, 2020 More than two-thirds 21 (68%) of the participants felt that they still needed further training with special mention on the identification and documentation of AEs. Less than a quarter 6 (19%) of the participants believed that the resources  Acceptability of the surveillance system in Zvimba District, 2020 All the participants 31 (100%) reported that they were the ones responsible for compiling AE data and expressed willingness to continue participating in the surveillance system and forwarding reports on VMMC adverse events monitoring to the next level. Acceptability of the surveillance system was also confirmed by the fact that all health facilities were compiling VMMC monthly return forms which were being submitted timeously to the next reporting level.
Simplicity of the surveillance system in Zvimba District, 2020 Only 10 (32%) of the participants reported that they had ever identified an adverse event following VMMC whilst 27 (87%) of the participants reported that they had never completed adverse events form. More than half of the participants 22 (71%) highlighted that it may take an average of 11 -20 minutes to complete the A/E reporting form and 16 (52%) of the participants regarded filling AE reporting form as difficult. The findings on the assessment of simplicity attribute of the surveillance system are highlighted in (Table 4).

Discussion
Overall, knowledge of study participants on the adverse events following VMMC surveillance system was graded as fair on the Likert Scale. was willing to continue participating in the surveillance system and timeously submitted surveillance data to the next level. The surveillance system was not simple as the majority of the study participants reported that they required additional training in documenting AE data and that they found the documentation process time-consuming. There were gaps in the adequate documentation of the adverse event monitoring as evidenced by some records not having data filled on all the days for the follow-up period. Lack of knowledge by study participants in describing criteria for severity and coding of adverse events could be attributed to the unavailability of displayed adverse event classification guidelines at some of the health care facilities and the perceived inadequate training on the surveillance system as highlighted by the majority of the study participants. In a similar study by Dube et al. (2017) possible reasons that were given by the study participants as affecting the simplicity of the surveillance system was the lack of periodic refresher courses on identifying and documenting adverse events and the general lack of knowledge on VMMC adverse events by health care workers [9]. In Zvimba District, refresher VMMC training workshops course for health care workers were now limited due to decreasing funding for the programme as the country is now moving into the sustainability phase for the VMMC programme that is characterized by less reliance on donor funding. Surveillance systems should be as simple as possible whilst still meeting their objectives as a simple system is more likely to provide timely data and is generally more acceptable to staff members thus improving participation as highlighted by Matambo et al. (2016) [10]. The lack of resources due to decreasing funding for the VMMC programme could have also affected the capacity for health facilities to conduct outreaches in hard-to-reach areas and thus affecting the coverage of the VMMC programme as a whole.
Findings from this study also showed that the surveillance system was also not flexible as no other changes have been made to accommodate any other adverse events. This is similar to findings by Dube et al. (2017) who found out that there were no adjustments that had been done in accommodating other adverse events in the surveillance system [9]. Flexibility is necessary when changes occur in case definitions, reporting forms, adding new health events and procedures. This will ensure that the surveillance system remains relevant in the face of an ever-evolving healthcare delivery system.
One of the objectives of this study was to ascertain the reasons that were contributing to possible underreporting of AE data as highlighted by study participants. Findings from the key informants' interview acknowledged that there could be possible under-reporting of adverse events following VMMC in Zvimba due to various weaknesses in the surveillance system. Similar findings were highlighted by Marongwe et al. (2019) who reported that in Zimbabwe, previous studies on data quality and AE identification within the male circumcision pro-gram revealed challenges on correct and complete data and that the low AE rate in Zimbabwe may indicate potential AE underreporting [11]. Gadzayi et al. (2020) reported that some of the reasons for underreporting in a study of adverse events following immunization in a study that was conducted in Manicaland province included fear by healthcare workers of being blamed for causing the adverse event, clients poor knowledge of adverse events which results in them not reporting at facilities when an A/E occurs and also the misclassification of events as mild [12]. Dube et al. (2017) state that reasons for under-reporting which are common include a surveillance system that does not capture marginalized high-risk groups, unawareness of the adverse events among health care workers and the clients not seeking medical care for various reasons. Clients do not seek medical care such as when they cannot take time off work, cannot access the hospital, or feel their illness does not warrant a trip to the hospital.

Limitations
We could not reach the minimum calculated sample size of 44 health care workers and this could have affected the power of the study.

Conclusions
From the records, we reviewed about a quarter of those records had incomplete data, especially for day 42 on the post-surgery monitoring of AEs. Health workers' knowledge was found to be fair on the three-point Likert scale but the majority scored poorly on describing the criteria used on classifying adverse events.
All these factors could have contributed to underreporting of adverse events and thus affected the accuracy of the rates being reported. The A/Es following VMMC surveillance system in Zvimba District was found to be acceptable, but unstable, not flexible and not representative.
We recommended retraining of all health care workers involved in VMMC on the adverse events following VMMC surveillance system to capacitate health care workers to be competent in identifying and documenting adverse events that may occur during the VMMC programme.

Funding
This study was funded by the AIDS and TB Program and Health Studies Office.