Tools and Manuals Development, Testing, and Ethical Clearance
Step 5
The determination of tools for data collection is driven by the data that are intended to be collected and made available by the SMSS. In consultation with the registration authorities, SMSS tools can be designed to accommodate all relevant variables for its CRVS integration. However, such tools must be field-tested prior its use in the SMSS platform. Data elements and tools in a mortality surveillance system are driven by the purpose of the system. The purpose can be the ascertainment of death counts, rates, age and cause specific distribution. In general, to be useful such system should not be limited to death counts but be able to produce meaningful and interpretable estimates of numerators and denominators for the mortality rates of interest. It would also include specific relevant sub-groupings to allow the assessment of disparities or inequalities.
Key Points
Essentials tools include a household population listing form, new household registration form, vital events recording forms (separate for pregnancy outcome and deaths), and verbal and social autopsy interviews form Manuals that explain the data collection process are needed to maintain standardized data collection across study clusters. Obtaining appropriate ethical clearance from the country should be considered.
Numerators for Mortality Rates
These are counts of deaths by sex, age and cause for the specific target group or geography. These counts can be ascertained through the recording of individual level information on date of death, age at death, date of birth, cause of death, and date of data collection. The ascertainment of cause of death must follow a rigorous process of diagnosis and classification of disease. To ensure comparison to other settings, the classification must be consistent with the global standard such the International Classification of Diseases (ICD). In low- and middle-income countries (LMIC), countries rely on medical certification of cause of death (MCCOD) for facility deaths. These consist of the review of the medical records by attending physician and completion of a death certificate. Given that most deaths in these countries occur outside facilities, verbal autopsies have become increasingly popular for ascertaining causes of death in the absence of clinical records with signs, symptoms and conditions surrounding the death. However, given the usually large uncertainty in verbal autopsy-based causes of death, there has not been a consensus on the use of its individual level cause diagnosis nor on the mixing of MCCOD-based cause of death with VA-based cause of death into a single dataset for analysis.
Denominators of Mortality Statistics
To be interpretable, mortality statistics are expressed in terms of rates for levels and trends assessment. Rates express relative frequency of deaths with a specific reference period and allow for comparison across geographies or specific groups. The typical denominators for mortality rates are person-time lived, calculated from total population by age and sex. Mortality rates can also be computed in terms of probabilities by converting typical rates to probabilities or directly calculating the probabilities by estimating the size of population at risk. A special case is stillbirth and childhood mortality that use births (total births or live births) to estimate probability of death. Other measures are expressed in terms of ratio, such as maternal mortality ratio. Thus, a surveillance system must identify a strategy for determining the number of births and the population size by age and sex. This information can be integrated within the surveillance system and collected from individual microdata and aggregated data. Data collection will include information on each population member, age, sex, and date of data collection. It also can be borrowed from other sources such as population census projections. However, these secondary sources do not always include the desired subgroups or disaggregation and can also suffer from quality issues.
Household Listing and Community Vital Events Recording Tools
The routine surveillance tools are the new household registration form and the vital events forms for pregnancy outcomes and deaths. A pregnancy surveillance form may be included if there is a desire to incorporate pregnancy surveillance in the SMSS. Details about these forms are below:
Household listing form — At the onset of the surveillance a list of the household population by age and sex is necessary to establish the denominators.
New household registration form — After the baseline population and household list is completed, the updated list of households should be printed and/or loaded into the mobile device and provided to the community workers. Community workers may be tasked with updating this list of households and members, including new households in their communities using the household registration form.
Vital event forms: pregnancy, pregnancy outcomes, and death — Pregnancies, births, and deaths are recorded as they occur. Pregnancy, birth, and death forms should include household identification, limited respondent demographics, and information regarding the person affected by the event, such as their name, age, sex, and date of the event.
Verbal and Social Autopsy (VASA) Forms
Causes of death are investigated through verbal autopsies interviews, which cover general information about households, as well as information on neonatal, child and adult deaths using specific questionnaires. The verbal autopsy forms are based on the standard tool developed and published by the World Health Organization (WHO).1 If there is a desire to also collect information on social autopsy of deaths, a social autopsy (SA) tool can be integrated into the verbal autopsy tool to provide more contextual information regarding the death. An integrated verbal and social autopsy tool has been developed and implemented as part of the Mozambique SMSS.2 Alternatively to the integrated verbal and social autopsy, a social autopsy tools can be used as optional module.
The Mozambique VASA questionnaire blends the 2016 WHO verbal autopsy questionnaire with the social autopsy questionnaire. All of the illness preventive and curative factors included in the SA questionnaires were derived from evidence–based interventions contained in the Lives Saved Tools (LiST) and judged by a Cochrane review for supportive evidence of efficacy or are among the newborn interventions recommended by the WHO. 3 In addition, where possible, the questions were worded similarly to those in the Demographic and Health Surveys in order to allow for the comparison of the SA data with similar data for survivors in settings where a recent DHS was conducted. 4
The VASA questionnaire is divided into four sub-questionnaires:
The general information questionnaire that pertains to all group ages. It includes sections on the background of the deceased, the VASA interview attempts, consent to conduct the interview and information about the respondent and other people present during the interview. The stillbirth and neonatal deaths questionnaire includes sections to further investigate any death in that group by the respondent if the baby moved, breathed, and cried at birth. When the answer was “no” to all items, the software assigned this case to the stillbirth category. This tool includes questions on maternal, preventive and curative care in the event of a neonatal death or stillbirth. The child (1 month to 11 years) questionnaire. The adult (12 years or older) questionnaire.
In these sub questionnaires, the verbal and social autopsy questions are chronologically blended, with social autopsy questions on preventive care asked First, then the verbal autopsy to identify the signs and symptoms of the illness that led to death, followed by social autopsy questions to assess the perception of the illness and care seeking carried out by the caregiver of the deceased person. The questionnaires are implemented using a tablet computer; it is important to program the software with skip patterns and internal consistency checks to improve the interview quality.
Manuals
In addition to providing standardized training, it is essential to have training manuals in place to maintain consistency during data collection. The following are descriptions of the necessary manuals.
Community Workers Manual
This manual describes tools and data collection procedures. It also provides information about: The roles and responsibilities of each team member from community workers, supervisors and coordinators; The ethical aspects of conducting research interviews, such as informed consent and confidentiality, effective communication and sensitivity issues while interacting with pregnant and young unmarried mothers, interview procedures and tips, and How to implement each question in the household, pregnancy, birth outcome and death forms.
VASA Manual
This manual describes interviewers’ roles and responsibilities, the supervisory procedures, ethical aspects of conducting research interviews, such as informed consent and confidentiality, effective communication and sensitivity issues while interacting with bereaved relatives of the deceased, interview procedures and tips, and provides a question-by-question guide to the VASA questionnaire. Interviewers are to use this manual as a reference guide during their training and during their work as VA or VASA interviewers as needed. The manual also serves as a reference for trainers and supervisors of the interviewers and for managers of the VASA project.
Open Data Kit (ODK) Manual
The Mozambique COMSA/SIS-COVE uses ODK software for electronic data collection (as described in step 6). This manual provides instructions on the ODK Collect main menu, how to navigate from one page to another, response types in interviews, common form messages, and subsections of an interview. It also provides information on ODK features that may assist interviewers during the data collection. Depending on the country’s adopted digital solutions, we recommend developing such a manual for the interviewers’ training, the data collection, and the reporting.
Field Supervision Manual
This manual describes activities that must take place prior to and during the field visits by VASA data collectors and supervisors, such as:
Central-level preparation that takes place before scheduling field visits, including procedures about data reported by the community workers. The number of households visited and the events reported in the past few months should be reviewed prior to the visit. This information is used to guide the selection of the communities to be visited for supervision.
Provincial-level preparation of scheduling the meeting with the provincial team leader and the VASA interviewers to prepare for the field visits.
Community-level preparation of meeting with community leaders to discuss the community worker’s activities and presenting basic information regarding events reported by the community workers, such as the total number of deaths and births in the past month or trimester. The supervision activities are conducted by visiting selected households to collect events and compare them with community workers’ events. At the end of the visit, the supervisor should meet with the interviewer to discuss any discrepancies and challenges in the field. If needed, the supervisor can recommend additional trainings for the interviewer.
Testing of Data Collection Tools
Data collection tools must be carefully reviewed and pilot tested to ensure that the questions are well understood and flow in logical order. Each tool will be programmed into a data capture software. The field pilot testing of the tool can be implemented following the training of data collectors, in order to integrate any corrections within the training of data collectors before deployment.
Ethical Clearance
Sample Mortality Surveillance System platforms should strictly follow ethical principles as outlined in the Declaration of Helsinki and its revisions and should be performed in compliance with the guidelines of Good Clinical Practice and Good Clinical Data Management Practice. 5 If not implemented by government as an extension of public health surveillance, the implementation of the SMSS will likely require obtaining ethical clearance from the relevant national institutional review board (IRB). A complete project protocol, including all tools, and informed consent, permission and assent forms must be put together according to the requirement of the national IRB.
Informed consent forms should be developed for each type of SMSS tools (household listing and registration, vital event registration, verbal and social autopsies).
In accordance with the ethical board, either an oral or written consent may be approved. The consent forms must be translated into the country’s official language and main local languages. The data collectors must be trained to carefully explain the study, read the informed consent form, and obtain approval from respondents before collecting any data. During the surveillance, it is possible that events such as pregnancy, birth or death of a child occur to a minor (<18 years). In these cases, if information on pregnancy, birth and cause of death must be collected, the data collector must obtain assent (assent form) from the minor participant and permission from the parent/guardian before the survey.
To Learn More
| Last updated |
|---|
| June 2, 2025 |