Enabling targeted mass drug administration for schistosomiasis in north-western Tanzania: Exploring the use of geostatistical modeling to inform planning at sub-district level.

INTRODUCTION: Schistosomiasis is a parasitic disease in Tanzania affecting over 50% of the population. Current control strategies involve mass drug administration (MDA) campaigns at the district level, which have led to problems of over- and under-treatment in different areas. WHO guidelines have called for more targeted MDA to circumvent these problems, however a scarcity of prevalence data inhibits decision makers from prioritizing sub-district areas for MDA. This study demonstrated how geostatistics can be used to inform planning for targeted MDA. METHODS: Geostatistical sub-district (ward-level) prevalence estimates were generated through combining a zero-inflated poisson model and kriging approach (regression kriging). To make predictions, the model used prevalence survey data collected in 2021 of 17,400 school children in six regions of Tanzania, along with several open source ecological and socio-demographic variables with known associations with schistosomiasis. RESULTS: The model results show that regression kriging can be used to effectively predict the ward level parasite prevalence of the two species of Schistosoma endemic to the study area. Kriging was found to further improve the regression model fit, with an adjusted R-squared value of 0.51 and 0.32 for intestinal and urogenital schistosomiasis, respectively. Targeted treatment based on model predictions would represent a shift in treatment away from 193 wards estimated to be over-treated to 149 wards that would have been omitted from the district level MDA. CONCLUSIONS: Geostatistical models can help to support NTD program efficiency and reduce disease transmission by facilitating WHO recommended targeted MDA treatment through provision of prevalence estimates where data is scarce.

Nodding Syndrome: Clinical Characteristics, Risks Factors, Access to Treatment, and Perceptions in the Greater Mundri Area, South Sudan.

We conducted a house-to-house survey in the Mundri, Western Equatoria state of South Sudan to investigate the clinical characteristics, risk factors, access to treatment and perceptions about nodding syndrome (NS). In total, 224 NS cases with median age of seizure onset of 10 years were identified. Head nodding only was reported in 50 (22.3%) cases, and head nodding plus other types of seizures in 174 (77.7%) cases. Wasting, stunted growth, delayed sexual development and speech and behavioral abnormalities were observed in 17 (23.6%), 16 (22.2%), 9 (17.3%), 14 (19.4%) and 4 (5.6%) cases, respectively. The consumption of rat meat, but not other bushmeat was associated with an increased risk of NS (OR 9.31, 95% CI 1.27-406.51). Children with NS were more likely to have taken ivermectin in the last 5 years (OR 2.40, 95% CI 1.33-4.43). NS cases were less likely to share a bedroom with other children (OR 0.06, 95% CI 0.02-0.16) or adults (OR 0.27, 95% CI 0.13-0.56). In conclusion, rat meat consumption is an unlikely risk factor for NS, and ivermectin intake was more common among NS cases than controls. Importantly, we documented that children with NS are stigmatized because of the misconception that NS is transmitted through direct contact.

Factors Affecting the Transition from Paper to Digital Data Collection for Mobile Tuberculosis Active Case Finding in Low Internet Access Settings in Pakistan.

Between September 2020 and March 2021, Mercy Corps piloted hybrid digital (CAPI) and paper-based (PAPI) data collection as part of its tuberculosis (TB) active case finding strategy. Data were collected using CAPI and PAPI at 140 TB chest camps in low Internet access areas of Punjab and Khyber Pakhtunkhwa provinces in Pakistan. PAPI data collection was performed primarily during the camp and entered using a tailor-performed CAPI tool after camps. To assess the feasibility of this hybrid approach, quality of digital records were measured against the paper "gold standard", and user acceptance was evaluated through focus group discussions. Completeness of digital data varied by indicator, van screening team, and month of implementation: chest camp attendees and pulmonary TB cases showed the highest CAPI/PAPI completeness ratios (1.01 and 0.96 respectively), and among them, all forms of TB diagnosis and treatment initiation were lowest (0.63 and 0.64 respectively). Vans entering CAPI data with high levels of completeness generally did so for all indicators, and significant differences in mean indicator completeness rates between PAPI and CAPI were observed between vans. User feedback suggested that although the CAPI tool required practice to gain proficiency, the technology was appreciated and will be better perceived once double entry in CAPI and PAPI can transition to CAPI only. CAPI data collection enables data to be entered in a more timely fashion in low-Internet-access settings, which will enable more rapid, evidence-based program steering. The current system in which double data entry is conducted to ensure data quality is an added burden for staff with many activities. Transitioning to a fully digital data collection system for TB case finding in low-Internet-access settings requires substantial investments in M&E support, shifts in data reporting accountability, and technology to link records of patients who pass through separate data collection stages during chest camp events.

Epidemiology of nodding syndrome in the Greater Mundri area, South Sudan: Prevalence, spatial pattern and environmental risk factors.

BACKGROUND: Nodding syndrome (NS) is a progressive neurological disease that has been described in several sub-Saharan African counties, but South Sudan is considered the most affected. However, knowledge about the exact burden and the epidemiological risk factors of NS in South Sudan is lacking. OBJECTIVE: To determine the prevalence, distribution and epidemiological risk factors of NS in the Greater Mundri area, the epicenter of NS in South Sudan. METHODS: A NS prevalence house-to-house survey was conducted in multiple villages between February 2018 and November 2019. Geographical distribution and clustering of NS cases was identified using spatial and binomial regression analysis. Epidemiological risk factors of NS were identified using univariate and multivariate models. RESULTS: Of the 22,411 persons surveyed in 92 villages, 607 (2.7%) persons with NS were identified, of which 114 (19%) were new-onset cases. The highest prevalence was found in Diko village with a prevalence of 13.7%. NS showed a significant spatial pattern with clustering of cases between adjacent households and along rivers. Risks factors for NS include all behaviors around rivers (drinking, cooking, handwashing and bathing) and exposure to poultry. On the other hand, ownership of mobile phone decreased the risk of NS. Many other factors, including prior ivermectin treatment and internal displacement were not associated with NS. CONCLUSION: Our study demonstrates a very high burden of the NS disease in the Greater Mundri area, strengthens the association with rivers, and identified possible new clues for an underlying cause.

TB Hackathon: Development and Comparison of Five Models to Predict Subnational Tuberculosis Prevalence in Pakistan.

Pakistan's national tuberculosis control programme (NTP) is among the many programmes worldwide that value the importance of subnational tuberculosis (TB) burden estimates to support disease control efforts, but do not have reliable estimates. A hackathon was thus organised to solicit the development and comparison of several models for small area estimation of TB. The TB hackathon was launched in April 2019. Participating teams were requested to produce district-level estimates of bacteriologically positive TB prevalence among adults (over 15 years of age) for 2018. The NTP provided case-based data from their 2010-2011 TB prevalence survey, along with data relating to TB screening, testing and treatment for the period between 2010-2011 and 2018. Five teams submitted district-level TB prevalence estimates, methodological details and programming code. Although the geographical distribution of TB prevalence varied considerably across models, we identified several districts with consistently low notification-to-prevalence ratios. The hackathon highlighted the challenges of generating granular spatiotemporal TB prevalence forecasts based on a cross-sectional prevalence survey data and other data sources. Nevertheless, it provided a range of approaches to subnational disease modelling. The NTP's use and plans for these outputs shows that, limitations notwithstanding, they can be valuable for programme planning.

Going digital: added value of electronic data collection in 2018 Afghanistan Health Survey.

BACKGROUND: Through a nationally representative household survey in Afghanistan, we conducted an operational study in two relatively secure provinces comparing effectiveness of computer-aided personal interviewing (CAPI) with paper-and-pencil interviewing (PAPI). METHODS: In Panjshir and Parwan provinces, household survey data were collected using paper questionnaires in 15 clusters, and OpenDataKit (ODK) software on electronic tablets in 15 other clusters. Added value was evaluated from three perspectives: efficient implementation, data quality, and acceptability. Efficiency was measured through financial expenditures and time stamped data. Data quality was measured by examining completeness. Acceptability was studied through focus group discussions with survey staff. RESULTS: Survey costs were 68% more expensive in CAPI clusters compared to PAPI clusters, due primarily to the upfront one-time investment for survey programming. Enumerators spent significantly less time administering surveys in CAPI cluster households (248 min survey time) compared to PAPI (289 min), for an average savings of 41 min per household (95% CI 25-55). CAPI offered a savings of 87 days for data management over PAPI. Among 49 tracer variables (meaning responses were required from all respondents), small differences were observed between PAPI and CAPI. 2.2% of the cleaned dataset's tracer data points were missing in CAPI surveys (1216/ 56,073 data points), compared to 3.2% in PAPI surveys (1953/ 60,675 data points). In pre-cleaned datasets, 3.9% of tracer data points were missing in CAPI surveys (2151/ 55,092 data points) compared to 3.2% in PAPI surveys (1924/ 60,113 data points). Enumerators from Panjsher and Parwan preferred CAPI over PAPI due to time savings, user-friendliness, improved data security, and less conspicuity when traveling; however approximately half of enumerators trained from all 34 provinces reported feeling unsafe due to Taliban presence. Community and household respondent skepticism could be resolved by enumerator reassurance. Enumerators shared that in the future, they prefer collecting data using CAPI when possible. CONCLUSIONS: CAPI offers clear gains in efficiency over PAPI for data collection and management time, although costs are relatively comparable even without the programming investment. However, serious field staff concerns around Taliban threats and general insecurity mean that CAPI should only be conducted in relatively secure areas.

Finding gaps in TB notifications: spatial analysis of geographical patterns of TB notifications, associations with TB program efforts and social determinants of TB risk in Bangladesh, Nepal and Pakistan.

BACKGROUND: In order to effectively combat Tuberculosis, resources to diagnose and treat TB should be allocated effectively to the areas and population that need them. Although a wealth of subnational data on TB is routinely collected to support local planning, it is often underutilized. Therefore, this study uses spatial analytical techniques and profiling to understand and identify factors underlying spatial variation in TB case notification rates (CNR) in Bangladesh, Nepal and Pakistan for better TB program planning. METHODS: Spatial analytical techniques and profiling was used to identify subnational patterns of TB CNRs at the district level in Bangladesh (N = 64, 2015), Nepal (N = 75, 2014) and Pakistan (N = 142, 2015). A multivariable linear regression analysis was performed to assess the association between subnational CNR and demographic and health indicators associated with TB burden and indicators of TB programme efforts. To correct for spatial dependencies of the observations, the residuals of the multivariable models were tested for unexplained spatial autocorrelation. Spatial autocorrelation among the residuals was adjusted for by fitting a simultaneous autoregressive model (SAR). RESULTS: Spatial clustering of TB CNRs was observed in all three countries. In Bangladesh, TB CNR were found significantly associated with testing rate (0.06%, p < 0.001), test positivity rate (14.44%, p < 0.001), proportion of bacteriologically confirmed cases (- 1.33%, p < 0.001) and population density (4.5*10-3%, p < 0.01). In Nepal, TB CNR were associated with population sex ratio (1.54%, p < 0.01), facility density (- 0.19%, p < 0.05) and treatment success rate (- 3.68%, p < 0.001). Finally, TB CNR in Pakistan were found significantly associated with testing rate (0.08%, p < 0.001), positivity rate (4.29, p < 0.001), proportion of bacteriologically confirmed cases (- 1.45, p < 0.001), vaccination coverage (1.17%, p < 0.001) and facility density (20.41%, p < 0.001). CONCLUSION: Subnational TB CNRs are more likely reflective of TB programme efforts and access to healthcare than TB burden. TB CNRs are better used for monitoring and evaluation of TB control efforts than the TB epidemic. Using spatial analytical techniques and profiling can help identify areas where TB is underreported. Applying these techniques routinely in the surveillance facilitates the use of TB CNRs in program planning.

Completeness of tuberculosis (TB) notification: inventory studies and capture-recapture analyses, six European Union countries, 2014 to 2016.

BackgroundProgress towards the World Health Organization's End TB Strategy is monitored by assessing tuberculosis (TB) incidence, often derived from TB notification, assuming complete case detection and reporting. This assumption is unlikely to hold in many settings, including European Union (EU) countries.AimWe aimed to assess observed and estimated completeness of TB notification through inventory studies and capture-recapture (CRC) methodology in six EU countries: Croatia, Denmark, Finland, the Netherlands, Portugal Slovenia.MethodsWe performed record linkage, case ascertainment and CRC analyses of data collected retrospectively from at least three national TB-related registers in each country between 2014 and 2016.ResultsObserved completeness of TB notification by inventory studies was 73.9% in Croatia, 98.7% in Denmark, 83.6% in Finland, 81.6% in the Netherlands, 85.8% in Portugal and 100% in Slovenia. Subsequent CRC analysis estimated completeness of TB notification to be 98.4% in Denmark, 76.5% in Finland and 77.0% in Portugal. In Croatia, CRC analyses produced implausible results while in the Netherlands and Slovenia, it was methodologically considered not meaningful.ConclusionInventory studies and CRC methodology suggest a TB notification completeness between 73.9% and 100% in the six EU countries. Mandatory reporting by clinicians and laboratories, and cross-checking of registers, strongly contributes to accurate notification rates, but hospital episode registers likely contain a considerable proportion of false-positive TB records and are thus less useful. Further strengthening routine surveillance to count TB cases, i.e. incidence, accurately by employing record-linkage of high-quality TB registers should make CRC studies obsolete in EU countries.

Getting more than "claps": incentive preferences of voluntary community-based mobilizers in Tanzania.

BACKGROUND: Marie Stopes Tanzania works with a voluntary cadre of 66 community-based mobilizers (CBMs), who are tasked with raising awareness, generating demand and providing referral to potential clients for family planning, comprehensive post-abortion care and cervical cancer screening. CBMs extend the reach of urban clinics to peri-urban communities, enhancing access to sexual and reproductive health services. In an effort to optimize performance of CBMs, a study was conducted to explore the drivers of CBM motivation and inform the design of an incentive scheme. METHODS: Three focus group discussions with 17 CBMs and 11 interviews with CBM supervisors and managers were conducted in three clinics and the head office. After thematic analysis of transcripts, findings on motivational factors were discussed in a reflection workshop and informed the development of a discrete choice experiment (DCE) involving 61 CBMs as respondents. The DCE included eight choice questions on two incentive schemes, each consisting of five attributes related to remuneration, training, supervision, benefits and identification. For each attribute, different incentive options were presented, based on the outcomes of the qualitative assessment. The DCE results were analysed using conditional logistic regression. RESULTS: A variety of factors motivated CBMs. Most CBMs were motivated to conduct their work because of an intrinsic desire to serve their community. The most mentioned extrinsic motivational factors were recognition from the community and supervisors, monthly allowance, availability of supporting materials and identification, trainings, supervision and feedback on performance. Recommendations for improvement were translated into the DCE. Incentive attributes that were found to be significant in DCE analysis (p < 0.05), in preference order, were carrying an ID card, bi-monthly training, supervision conducted via both monthly meetings at clinics and visits from the head office, and a monthly flat rate remuneration (over pay for performance). CONCLUSION: Despite the recognition that being a CBM is voluntary, incentives, especially those of non-financial nature, are important motivators. Incentive schemes should include basic compensation with a mix of other incentives to facilitate CBMs' work and enhance their motivation. Programme designs need to take into account the voices of community-based workers, to optimize their performance and service delivery to communities they serve.

Development and validation of a predictive ecological model for TB prevalence.

Background: Nationally representative tuberculosis (TB) prevalence surveys provide invaluable empirical measurements of TB burden but are a massive and complex undertaking. Therefore, methods that capitalize on data from these surveys are both attractive and imperative. The aim of this study was to use existing TB prevalence estimates to develop and validate an ecological predictive statistical model to indirectly estimate TB prevalence in low- and middle-income countries without survey data. Methods: We included national and subnational estimates from 30 nationally representative surveys and 2 district-level surveys in India, resulting in 50 data points for model development (training set). Ecological predictors included TB notification and programmatic data, co-morbidities and socio-environmental factors extracted from online data repositories. A random-effects multivariable binomial regression model was developed using the training set and was used to predict bacteriologically confirmed TB prevalence in 63 low- and middle-income countries across Africa and Asia in 2015. Results: Out of the 111 ecological predictors considered, 14 were retained for model building (due to incompleteness or collinearity). The final model retained for predictions included five predictors: continent, percentage retreated cases out of all notified, all forms TB notification rates per 100 000 population, population density and proportion of the population under the age of 15. Cross-fold validations in the training set showed very good average fit (R-sq = 0.92). Conclusion: Predictive ecological modelling is a useful complementary approach to indirectly estimating TB burden and can be considered alongside other methods in countries with limited robust empirical measurements of TB among the general population.