Welcome to SIOP 2022 Interactive Programme
The Congress will officially run on CET time zone (Central European Time, Barcelona)
- Joyce B. Kambugu (Uganda)
- Leila Hessissen (Morocco)
TYPE OF CANCER AND ALTERNATIVE MEDICINE DETERMINE PATIENT DELAY EXPERIENCED BY CHILDREN WITH CANCER: A STUDY IN WEST JAVA INDONESIA
- Nur M. Sari (Indonesia)
Abstract
Background and Aims
The majority of patients with pediatric cancer in developing countries present at an advanced stage due to delayed diagnostics, which has been recognized as an important barrier to effective care. The objective of the present study was to determine the influence of some clinical and social factors associated with patient delay.
Methods
This cross-sectional study was conducted at Dr. Hasan Sadikin Hospital, using data retrieved from the Indonesian Pediatric Cancer Registry (IP-Car) (2020–2021) for clinical variables, completed with interviews with the parents that used structured questionnaires to obtain data on their sociodemographic characteristics. Patient delay was defined as the number of days between the onset of the first symptoms associated with cancer, and the date the patient came to the health facility for the first time. presented using the median with IQR and range (minimum and maximum,). Binary logistic regression analysis model was fitted to identify factors associated with patient delay that classified through its median. A p-value of ≤ 0.05 was considered to indicate a statistically significant association.
Results
We interviewed the parents of 356 children with cancer. The median patient delay was 14 days (IQR 6–46.5 days. The most extended delay was in patients with malignant bone tumors (median 66, IQR: 14–126). In the multivariable logistic regression analysis, solid cancer [OR =5.22, 95% confidence interval (CI): 2.79–9.77, P <0.001], and the use of alternative medicine (OR =1.86, 95% CI: 1.13–3.08, P =0.015) were associated with patient delay.
Conclusions
Type of cancer and use of alternative medicine are essential factors that cause patient delay more than age and other family socioeconomic factors. Increasing public knowledge and awareness, and improving access to care should be improved, especially in a diagnosis of solid tumors.
IMPLEMENTING THE PEDIATRIC ONCOLOGY FACILITY INTEGRATED LOCAL EVALUATION (PROFILE): LESSONS LEARNT FROM OUR FACILITY BETA TESTING EXPERIENCE IN JORDAN
- Rawad Rihani (Jordan)
Abstract
Background and Aims
PrOFILE is a collaborative experience that aims to help pediatric hematology-oncology (PHO) teams to define an institutional improvement strategy and build local capacity in quality improvement (QI). In July 2021, the King Hussein Cancer Center (KHCC) team joined the second Full PrOFILE Beta Testing cohort. We aim to describe our experience, and lessons learnt implementing PrOFILE.
Methods
PrOFILE was implemented in three phases: preparation, assessment, and interpretation and action. We recruited our assessment team and selected three providers to complete QI Basic Certification during the preparation phase. Then, we collected the objective and subjective data for the twelve PrOFILE modules (26 forms). Objective data was collected by the Site Coordinator (SC) and reviewed and approved by the Physician Lead (PL). Subjective data was entered directly by a pool of Point of Care (POC) staff. We conducted six QI exercises and provided data validation and feedback for five short-interval reports. Finally, we used our score-based and descriptive reports to plan our local workshop and develop our institution's action plan.
Results
A total of 28 interdisciplinary healthcare providers formed the assessment team. Threeselected members completed the QI Basic Certification. PL and SC attended a total of 33 mentoring sessions and completed 14PrOFILE educational videos. Our form completion rate was 100% for both objective and subjective data. KHCC site scored 74% for Context, 92% for Workforce, 94% for Diagnostic, 94% for Therapy, and 100% for Patient and Outcome components. We found discrepancies in awareness level on institutional planning between POC staff (42%,range:15-100%) and PL and SC (100%), mainly in the National context and Finances and Resources modules.
Conclusions
PrOFILE implementation was feasible at our institution. Having our institutional leaders' support was essential. The tool helped us identify our strengths and prioritize opportunities amenable for improvement. Multisite PrOFILE implementation is valuable for regional and international benchmarking.
SOCIO-ECONOMIC STATUS IS A DETERMINANT OF CHILDHOOD CANCER OUTCOMES IN SOUTH AFRICAN CHILDREN
- Marc Hendricks (South Africa)
Abstract
Background and Aims
Significantly discrepant survival rates have been documented in single disease childhood cancer cohorts in South Africa, in which those from higher socioeconomic groups were shown to have a significantly lower risk of death than those from less affluent households. This study aimed to determine the impact of socioeconomic status (SES) on survival of children with cancer using pooled South African data.
Methods
Five databases spanning January 2000 to December 2021 were interrogated. SES status was assigned based on a public sector annual household income classification. H0 households (formally unemployed) received free healthcare. H1, H2 and H3 households paid relative to income. H3 households (highest income) earned more than USD24,000 per year. The Spearman test was used to assess correlations between SES and disease stage at diagnosis. Hazard ratios were determined using Cox regression modelling. The Kaplan-Meier procedure was used to estimate overall survival (OS) (CI 95%).
Results
Sixteen-hundred patients were eligible for analysis while 1,269 solid tumour patients were compared for SES category and stage. There was a negative correlation between SES and stage (Spearman rho= -0.178; p<0.001). Patients with solid tumours and lower SES showed proportionately higher numbers of stage III and IV disease (p<0.01). This proportion decreased with higher SES categories. In the multivariate analyses adjusted for sex, age, tumour type and stage, higher SES was associated with a lower risk of death (p<0.001), indicating that the impact of SES on survival was in excess of any effect that could be explained by lower stage disease alone. Five-year OS was 85.3% in children from H3 households versus 47.7% in children from H0 households (p<0.001).
Conclusions
SES significantly impacts cancer survival in South African children. Advocacy to increase social support for the poor is essential if we are to achieve equitable improvements in childhood cancer outcomes with standardised national protocols.
CANCER INCIDENCE AMONG CHILDREN IN UGANDA: A RETROSPECTIVE POPULATION-BASED STUDY, 2013-2017
- Annet Nakaganda (Uganda)
Abstract
Background and Aims
Cancer is a leading cause of childhood mortality in sub-Saharan Africa (SSA). Cancer control requires accurate estimates of the burden however, estimates for children in the region are based on only six national cancer registries, covering about 5% of Africa’s children population. This analysis was conducted to estimate cancer incidence and identify geographic variations among children with cancer in Uganda.
Methods
From 2018 to 2020, with a retrospective catchment population approach, we examined data from three regions (Central, Western and Eastern). Five districts from each region were selected using stratified random sampling. All newly diagnosed cancer cases from routine medical records of all health facilities serving the study populations in the period 2013-2017 were included. Cases were coded according to the International Classification of Diseases for Oncology (ICD-0-03). Data was analysed using CanReg5.
Results
A total of 718 children (0-14 years) with cancer were registered (390 males, 328 females). The overall age-standardized incidence rate (ASIR) was 59.5 per million children. Most common cancers were lymphomas (ASIR of 15.6), soft tissue sarcomas (ASIR of 9.8), retinoblastoma (ASIR of 8.4), leukemia (ASIR of 7.0) and renal tumors (ASIR of 6.6). Within the lymphoma group, 40% were Burkitt lymphomas and in the soft tissue sarcomas group, 46 % were rhabdomyosarcoma and 43% were Kaposi sarcoma. Overall ASIRs varied from 123.5 per 1,000,000 in Kampala capital city to 56.6 in central region; 41.5 in western region; and 30.5 in Eastern region.
Conclusions
The incidence rates varied greatly among regions which may reflect existing disparities in children’s cancer services/diagnosis in Uganda. This study validates a retrospective catchment population approach to improving cancer surveillance and the results will contribute to the development of national childhood cancer registration, and provide evidence for improving national cancer control in Uganda.
ACCURATELY FORECASTING PEDIATRIC HEMATOLOGY AND ONCOLOGY PATIENT VOLUMES AT TREATMENT CENTERS IN SUB-SAHARAN AFRICA
- Mark Zobeck (United States of America)
Abstract
Background and Aims
Forecasting the number of new patients at pediatric hematology-oncology (PHO) treatment centers in resource-constrained settings is important to optimize care delivery but can be difficult due to the month-to-month variability of the counts and changes in the expected number over time. We conducted a forecasting project to predict the number of new hematology and oncology patients at PHO treatment centers in Botswana (Botswana Treatment Center; BTC), Malawi (MTC), and Uganda (UTC) in 2021.
Methods
Monthly counts of new hematology and oncology patients were obtained between October 2016–March 2021. Time series models were fit to the data from Oct 2016-Dec 2020. Moving average and linear trend models were fit to the data with different time periods. The models were scored with validation data from January-March 2021 according to mean average error, root mean squared error, and lag-1 autocorrelation. The models with the top scores were chosen to forecast the mean and 95% prediction interval for Apr-Dec 2021. Differences between the observed and mean predicted (O-P) numbers of patients were calculated.
Results
Linear trend models performed the best for all sites except for oncology patients at UTC where a moving average model that excluded the period of COVID-19 scored the best. The O-P differences for new oncology patients was 0(O-P;45-45) for BTC, +21(129-107) for MTC, and +14(186-172) for UTC. The difference for new hematology patients was +2(45-43) for BTC, +47(159-112) for MTC, -12(607-619) for UTC. The total forecast error for all sites was +71(1171-1100), or 6.1% of the total patients. All observed counts fell within the 95% prediction intervals except new hematology patients at MTC which fell outside the upper limit.
Conclusions
Forecasting models from operations data can predict the total newly diagnosed patients at PHO treatment centers within an error of 6.1% of the observed total number of patients.
ANALYSIS OF CHILDHOOD CANCER MEDICINE ACCESS IN FOUR EAST AFRICAN COUNTRIES
- Avram Denburg (Canada)
Abstract
Background and Aims
Reliable access to essential childhood cancer medicines (CCMs) is key to improved childhood cancer outcomes in low- and middle-income countries (LMICs). We conducted an analysis of CCM access in four East African countries – Kenya, Rwanda, Tanzania and Uganda – to determine the availability and cost of essential cancer medicines and impacts of medicine unavailability on treatment interruption for five common pediatric malignancies.
Methods
We analyzed costs and stockout days for 34 essential drugs, based on 12 months of prospective price and inventory data from eight institutions across the study jurisdictions. SIOP adapted-treatment regimens were employed to assess the impact of stockouts on treatment interruptions. A mixed-effects logistic regression model examined associations between price, procurement efficiency (as median price ratio), site and drug availability.
Results
Stockouts for many cytotoxic and supportive care medicines were observed across sites, with highest mean unavailability in Kenya (49%), Rwanda (39%), and Tanzania (32%). Medicines vulnerable to stockouts across ≥ 4 sites included: methotrexate, bleomycin, etoposide, ifosfamide, oral morphine, and allopurinol. Although our analysis revealed a statistically significant difference in MPR across sites (range 0.399-1.28, p<0.01), average MPR at each site was below the internationally accepted threshold for efficient procurement (MPR < 1.5). No significant association between MPR and unavailability was found. Impacts of stockouts on treatment were noted across the majority of sites, with greatest potential for treatment interruptions in Hodgkin lymphoma, retinoblastoma and acute lymphoblastic leukemia.
Conclusions
Our findings provide detailed evidence of drug cost and availability in East Africa with implications for CCM access in the region. These data can inform support national and regional policymaking to optimize cancer drug availability and affordability as part of efforts to improve childhood cancer outcomes in the region.