Artificial intelligence for strengthening healthcare systems in low- and middle-income countries: a systematic scoping review.

In low- and middle-income countries (LMICs), AI has been promoted as a potential means of strengthening healthcare systems by a growing number of publications. We aimed to evaluate the scope and nature of AI technologies in the specific context of LMICs. In this systematic scoping review, we used a broad variety of AI and healthcare search terms. Our literature search included records published between 1st January 2009 and 30th September 2021 from the Scopus, EMBASE, MEDLINE, Global Health and APA PsycInfo databases, and grey literature from a Google Scholar search. We included studies that reported a quantitative and/or qualitative evaluation of a real-world application of AI in an LMIC health context. A total of 10 references evaluating the application of AI in an LMIC were included. Applications varied widely, including: clinical decision support systems, treatment planning and triage assistants and health chatbots. Only half of the papers reported which algorithms and datasets were used in order to train the AI. A number of challenges of using AI tools were reported, including issues with reliability, mixed impacts on workflows, poor user friendliness and lack of adeptness with local contexts. Many barriers exists that prevent the successful development and adoption of well-performing, context-specific AI tools, such as limited data availability, trust and evidence of cost-effectiveness in LMICs. Additional evaluations of the use of AI in healthcare in LMICs are needed in order to identify their effectiveness and reliability in real-world settings and to generate understanding for best practices for future implementations.

Surgical interventions for bilateral congenital cataract in children aged two years and under.

BACKGROUND: Congenital cataracts are lens opacities in one or both eyes of babies or children present at birth. These may cause a reduction in vision severe enough to require surgery. Cataracts are proportionally the most treatable cause of visual loss in childhood, and are a particular problem in low-income countries, where early intervention may not be possible. Paediatric cataracts provide different challenges to those in adults. Intense inflammation, amblyopia (vision is obstructed by cataract from birth which prevents normal development of the visual system), posterior capsule opacification and uncertainty about the final trajectory of ocular growth parameters can affect results of treatment. Two options currently considered for children under 2 years of age with bilateral congenital cataracts are: (i) intraocular lens (IOL) implantation; or (ii) leaving a child with primary aphakia (no lens in the eye), necessitating the need for contact lenses or aphakic glasses. Other important considerations regarding surgery include the prevention of visual axis opacification (VAO), glaucoma and the route used to perform lensectomy. OBJECTIVES: To assess the effectiveness of infant cataract surgery or lensectomy to no surgery for bilateral congenital cataracts in children aged 2 years and under. SEARCH METHODS: We searched the Cochrane Central Register of Controlled Trials (CENTRAL; which contains the Cochrane Eyes and Vision Trials Register; 2022, Issue 1); Ovid MEDLINE; Ovid Embase; the ISRCTN registry; ClinicalTrials.gov and the WHO ICTRP. The date of the search was 25 January 2022. SELECTION CRITERIA: We included all randomised controlled trials (RCTs) that compared infant cataract surgery or lensectomy to no surgery, in children with bilateral congenital cataracts aged 2 years and younger. This update (of a review published in 2001 and updated in 2006) does not include children over 2 years of age because they have a wider variety of aetiologies, and are therefore managed differently, and have contrasting outcomes. DATA COLLECTION AND ANALYSIS: We used standard methods expected by Cochrane. Two review authors extracted data independently. We assessed the risk of bias of included studies using RoB 1 and assessed the certainty of the evidence using GRADE. MAIN RESULTS: We identified three RCTs that met our inclusion criteria with each trial comparing a different aspect of surgical intervention for this condition. The trials included a total of 79 participants under 2 years of age, were conducted in India and follow-up ranged from 1 to 5 years. Study participants and outcome assessors were not masked in these trials. One study (60 children) compared primary IOL implantation with primary aphakia. The results from this study suggest that there may be little or no difference in visual acuity at 5 years comparing children with pseudophakia (mean logMAR 0.50) and aphakia (mean logMAR 0.59) (mean difference (MD) -0.09 logMAR, 95% confidence intervals (CIs) -0.24 to 0.06; 54 participants; very low-certainty evidence), but the evidence is very uncertain. The evidence is very uncertain as to the effect of IOL implantation compared with aphakia on visual axis opacification (VAO) (risk ratio (RR) 1.29, 95% CI 0.23 to 7.13; 54 participants; very low-certainty evidence). The trial investigators did not report on the cases of amblyopia. There was little evidence of a difference betwen the two groups in cases of glaucoma at 5 years follow-up (RR 0.86, 95% CI 0.24 to 3.10; 54 participants; very low-certainty evidence). Cases of retinal detachment and reoperation rates were not reported. The impact of IOL implantation on adverse effects is very uncertain because of the sparse data available: of the children who were pseudophakic, 1/29  needed a trabeculectomy and 8/29 developed posterior synechiae. In comparison, no trabeculectomies were needed in the aphakic group and 2/25 children had posterior synechiae (54 participants; very low-certainty evidence).  The second study (14 eyes of 7 children under 2 years of age) compared posterior optic capture of IOL without vitrectomy versus endocapsular implantations with anterior vitrectomy (commonly called 'in-the-bag surgery'). The authors did not report on visual acuity, amblyopia, glaucoma and reoperation rate. They had no cases of VAO in either group. The evidence is very uncertain as to the effect of in-the-bag implantation in children aged under 1 year. There was a higher incidence of inflammatory sequelae: 4/7 in-the-bag implantation eyes and 1/7 in optic capture eyes (P = 0.04, 7 participants; very low-certainty evidence). We graded the certainty of evidence as low or very low for imprecision in all outcomes because their statistical analysis reported that a sample size of 13 was needed in each group to achieve a power of 80%, whereas their subset of children under the age of 1 year had only 7 eyes in each group. The third study (24 eyes of 12 children) compared a transcorneal versus pars plana route using a 25-gauge transconjunctival sutureless vitrectomy system. The evidence is very uncertain as to the effect of the route chosen on the incidence of VAO, with no cases reported at 1 year follow-up in either group. The investigators did not report on visual acuity, amblyopia, glaucoma, retinal detachment and reoperation rate. The pars plana route had the adverse effects of posterior capsule rupture in 2/12 eyes, and 1/12 eyes needing sutures. Conversely, 1/12 eyes operated on by the transcorneal route needed sutures. We graded the outcomes with very low-certainty because of the small sample size and the absence of a priori sample size calculation. AUTHORS' CONCLUSIONS: There is no high level evidence for the effectiveness of one type of surgery for bilateral congenital cataracts over another, or whether surgery itself is better than primary aphakia. Further RCTs are required to inform modern practice about concerns, including the timing of surgery, age at which surgery should be undertaken, age for implantation of an IOL and development of complications, such as reoperations, glaucoma and retinal detachment. Standardising the methods used to measure visual function, along with objective monitoring of compliance with the use of aphakic glasses/contact lenses would greatly improve the quality of study data and enable more reliable interpretation of outcomes.

Back to the future? Lessons from the history of integrated child health services in England.

The UK has a long history of attempts to integrate child health services to improve outcomes, an ambition renewed in the recent The NHS Long Term Plan. It's therefore timely to review the history of integration to inform future initiatives. Key milestones include the Platt report (1959), Court report (1976), Sure Start (1999), National service framework (2004) and Facing the future report (2015). These stand against a backdrop of national NHS policy changes, with a myriad of local integration initiatives and research efforts in parallel. We suggest lessons for future integration initiatives: integration may support the quadruple aim; integration depends on addressing divides between primary and secondary care; workforce and funding challenges need to be resolved before integration can thrive; high-quality research and evaluation of integrated interventions is required; strong relationships between professional groups are key to integration; and integration can help address health inequalities.