Division of Paediatric Haematology/Oncology, The Hospital for Sick Children, Toronto, Ontario, Canada.
Cameroon Baptist Convention Hospitals, Mbingo, Cameroon.
Pediatr Blood Cancer. 2020 Dec;67(12):e28761. doi: 10.1002/pbc.28761. Epub 2020 Oct 10.
Childhood cancer outcomes in low- and middle-income countries (LMICs) lag behind those in high-income countries (HICs), in part due to late presentation and diagnosis. Though several interventions targeting early detection of childhood cancer have been implemented in LMICs, little is known about their efficacy.
We conducted a systematic review to identify studies describing such interventions. We searched multiple databases from inception to December 4, 2019. Studies were included if they reported on LMIC interventions focused on: (a) training of health care providers on early recognition of childhood cancer, or (ii) public awareness campaigns. We used preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines to conduct our review. The risk of bias in nonrandomized studies of interventions (ROBINS-I) checklist was used to assess quality of studies.
Twelve studies met inclusion criteria (n = 5 full text, n = 7 abstract only). Five studies focused on retinoblastoma only, while the others focused on all types of childhood cancer. The majority studied multiple interventions of which early detection was one component, but reported overall outcomes. All identified studies used pre-post evaluative designs to measure efficacy. Five studies reported statistically significant results postintervention: decrease in extraocular spread of retinoblastoma, decrease in rates of refusal/abandonment of treatment, increase in number of new referrals, increase in knowledge, and an absolute increase in median 5-year survival. Other studies reported improvements without tests of statistical significance. Two studies reported no difference in survival postintervention. The ROBINS-I checklist indicated that all studies were at serious risk of bias.
Though current evidence suggests that LMIC interventions targeting early detection of childhood cancer through health professional training and/or public awareness campaigns may be effective, this evidence is limited and of poor quality. Robust trials or quasi-experimental designs with long-term follow up are needed to identify the most effective interventions. Such studies will facilitate and inform the widespread uptake of early detection interventions across LMIC settings.
儿童癌症在中低收入国家（LMICs）的结局落后于高收入国家（HICs），部分原因是晚期表现和诊断。尽管在 LMICs 中已经实施了几项针对儿童癌症早期发现的干预措施，但对其疗效知之甚少。
我们进行了系统评价，以确定描述这些干预措施的研究。我们从开始到 2019 年 12 月 4 日搜索了多个数据库。如果研究报告了针对以下方面的 LMIC 干预措施，则将其纳入：（a）对医疗保健提供者进行儿童癌症早期识别的培训，或（ii）公众宣传运动。我们使用系统评价和荟萃分析的首选报告项目（PRISMA）指南来进行审查。非随机干预研究的偏倚风险（ROBINS-I）检查表用于评估研究质量。
符合纳入标准的 12 项研究（n=5 篇全文，n=7 篇摘要）。五项研究仅集中于视网膜母细胞瘤，而其他研究则集中于所有类型的儿童癌症。大多数研究了多种干预措施，其中早期检测是一个组成部分，但报告了总体结果。所有确定的研究都使用预-后评估设计来衡量疗效。五项研究报告了干预后有统计学意义的结果：减少视网膜母细胞瘤的眼外扩散，降低拒绝/放弃治疗的比率，增加新转诊人数，增加知识，以及绝对增加中位 5 年生存率。其他研究报告了没有统计学意义的改善。两项研究报告干预后生存率无差异。ROBINS-I 检查表表明，所有研究都存在严重的偏倚风险。
尽管目前的证据表明，通过医疗专业人员培训和/或公众宣传运动针对儿童癌症早期发现的 LMIC 干预措施可能是有效的，但这些证据有限且质量较差。需要进行稳健的试验或具有长期随访的准实验设计，以确定最有效的干预措施。此类研究将促进和为在 LMIC 环境中广泛采用早期检测干预措施提供信息。
