Implementation factors of non-communicable disease policies and programmes for children and youth in low-income and middle-income countries: a systematic review

Abstract

Introduction

Despite declared life-course principles in non-communicable disease (NCD) prevention and management, worldwide focus has been on older rather than younger populations. However, the burden from childhood NCDs has mounted; particularly in low-income and middle-income countries (LMICs). There is limited knowledge regarding the implementation of paediatric NCD policies and programmes in LMICs, despite their disproportionate burden of morbidity and mortality. We aimed to understand the barriers to and facilitators of paediatric NCD policy and programme implementation in LMICs.

Methods

We systematically searched medical databases, Web of Science and WHOLIS for studies on paediatric NCD policy and programme implementation in LMICs. Screening and quality assessment were performed independently by researchers, using consensus to resolve differences. Data extraction was conducted within the WHO health system building-blocks framework. Narrative thematic synthesis was conducted.

Results

93 studies (1992–2020) were included, spanning 86 LMICs. Most were of moderate or high quality. 78% reported on paediatric NCDs outside the four major NCD categories contributing to the adult burden. Across the framework, more barriers than facilitators were identified. The most prevalently reported factors were related to health service delivery, with system fragmentation impeding the continuity of age-specific NCD care. A significant facilitator was intersectoral collaborations between health and education actors to deliver care in trusted community settings. Non-health factors were also important to paediatric NCD policies and programmes, such as community stakeholders, sociocultural support to caregivers and school disruptions.

Conclusions

Multiple barriers prevent the optimal implementation of paediatric NCD policies and programmes in LMIC health systems. The low sociopolitical visibility of paediatric NCDs limits their prioritisation, resulting in fragmented service delivery and constraining the integration of programmes across key sectors impacting children, including health, education and social services. Implementation research is needed to understand specific contextual solutions to improve access to paediatric NCD services in diverse LMIC settings.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• The mortality burden of paediatric non-communicable diseases (NCDs) is disproportionately high in low-income and middle-income countries (LMICs).

• Compared with policies and programmes in paediatric communicable diseases and adult NCDs in LMICs, few systematic studies of those for paediatric NCDs have been done.

WHAT THIS STUDY ADDS

• Paediatric NCDs lie mostly in the ‘other’ category of the conventional ‘5×5’ framing of NCD agendas, lending to their low visibility.

• Health system fragmentation is a key barrier to implementing programmes in which children receive timely, age-specific care.

• Because of children’s dependence on their families and communities, non-health sector factors have important impacts on successful implementation of effective NCD policies and programmes, for example, the task-shifting of NCD care to education workers.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• A reframing of the ‘5×5’ international agenda for tackling NCDs is required to bring more focus to paediatric NCDs, both in policymaking and in research.

• The younger demographic profile of LMICs suggests that unless paediatric NCDs are adequately addressed, adult NCDs will become an even greater problem, since about 70% of premature deaths due to NCDs have their roots in risks established early in life.

• The distinctiveness of children’s needs underscores the importance of committing to child-relevant implementation research that encompasses the sociocultural environment of NCD policies and programmes, extending beyond the bounds of the health system.

Introduction

The management of non-communicable diseases (NCDs) has become a major public health focus worldwide. Despite the declared all-ages and life-course principles in international initiatives, there has been a disproportionate focus of NCD policies and programmes among older populations due to their higher prevalence. NCDs, however, have a profound impact on children and youth. As childhood mortality and morbidity associated with communicable diseases have diminished, the burden from NCDs has mounted.

NCDs are chronic conditions resulting from genetic, physiological, environmental and behavioural influences, rather than person-to-person transmission.^{1 2} Although the definition remains consistent across income levels and age groups, the patterns of prevalence and burden vary significantly. Globally, 1 in 10 youth live with a physical disability, with 80% residing in low-income and middle-income countries (LMICs).² This statistic does not specify whether the results stem from communicable diseases, or NCDs, particularly injuries. As such, paediatric NCDs within LMICs necessitate comprehensive scope, inclusive of grey areas such as physical and congenital disabilities, as to not limit burden characterisation.

An estimated 1 million paediatric deaths (aged 0–19 years) were attributed to NCDs in 2017, and incident cases numbered about 250 million.² Prevalence of the major NCDs, in millions, was estimated at: mental health disorders—231.3; injuries and violence—170.4; chronic respiratory disorders—108.9; cardiovascular disease—13.9; diabetes—8.8 and cancer—5.9.² Among adolescents aged 10–19 years, about 20% of deaths in 2019 were attributable to NCDs.³ Childhood behavioural disorders comprised the highest combined mortality/morbidity toll in those aged 10–14 years, with 433 disability-adjusted life-years (DALYs) lost per 100 000 population. Depressive disorders exacted the highest toll among those aged 15–19 years, with a loss of 497 DALYs/100 000.³

Moreover, the health behaviours that drive the development of NCDs in adulthood, such as tobacco and alcohol use, physical inactivity and unhealthy eating, often have their beginnings in childhood and adolescence. Akseer et al estimate that 80% of boys and girls aged 11–17 across all regions of the world have insufficient physical activity.⁴

As for many other diseases and conditions, the mortality due to paediatric NCDs in LMICs is disproportionately higher than in high-income countries (HICs). Lower GDP and health expenditure per capita are strong predictors of higher NCD burden among adults and adolescents.⁴ Approximately three-quarters of child deaths attributable to NCDs take place in LMICs, but this is likely an underestimate, as surveillance for and monitoring of childhood NCDs is limited in low-resource settings.²

In response, policies, regulations and programmes addressing NCDs in children and youth have been implemented in LMICs. For example, in the WHO Africa region, national strategies for health/family planning exist in 95% of countries; mental health, 82%; nutritional interventions, 80% and tobacco control, 79%, among others. Yet, the implementation status of these strategies remains largely unknown, as does the understanding of the challenges encountered and the successes achieved in their execution.⁴

To advance understanding of LMIC health system responses to paediatric NCDs, we conducted a systematic review to describe policy and programme implementation and to identify barriers to and facilitators of successful implementation.

Methods

Search strategy

This systematic review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist.⁵ We systematically searched MEDLINE, EMBASE Classic, EMBASE and the Cochrane Central Register of Controlled Trials, Web of Science and WHOLIS from database inception to 23 July 2020. Search concepts included NCDs, policy and programme implementation or evaluation, LMICs and children, infants or adolescents (sample search strategy in online supplemental appendix A).

Screening and study selection

Inclusion criteria were: (a) addressed policy or programme implementation; (b) reported on an LMIC setting, based on World Bank classifications; (c) focused on one or more NCDs and (d) included paediatric populations (<18 years).⁶ We excluded editorials, commentaries, opinion pieces, formal systematic reviews, study protocols, studies with narrow clinical focus, studies on injuries, studies where no specific health intervention was described (eg, prevalence studies, cost-effective analyses), non-peer-reviewed published proceedings, conference abstracts and non-English language studies. Abstracts were imported into Covidence, and duplicates were removed. Titles and abstracts were independently screened by two of three reviewers (LW, NM, KP). Discrepancies were discussed and resolved by joint full article review, and a fourth reviewer (AD) facilitated consensus when necessary. Two out of six reviewers (NM, GTN, LW, JN, BW, KP) independently screened full texts, and discrepancies were resolved by joint reviews. Two additional reviewers (CG, AD) helped reach consensus.

Data extraction and quality assessment

A data extraction form was developed and piloted by NM (online supplemental appendix B). GN independently reviewed and extracted relevant data from each study, including, year of publication; NCD type; continuum-of-care component; policy/programme implementation stage, platform and system level; target population and geographic area. Author-reported health system barriers and facilitators were extracted and mapped to the six WHO health system building-blocks and an additional non-health systems ‘other’ block, to enable a comprehensive health system perspective.⁷ Two reviewers (GTN, NM) independently assessed the quality of each included article, using the Mixed Methods Appraisal Tool and the Joanna Briggs Institute critical appraisal checklist for systematic reviews and research syntheses.^{8 9} Discrepancies were resolved by consensus.

Data synthesis

A narrative synthesis was conducted by classifying barriers and facilitators using descriptive codes, then grouping these codes into broader categories or themes. We descriptively analysed and generated frequency plots, maps and tables for the following characteristics: year of publication, geographic distribution of studies, NCD classifications, continuum-of-care component (ie, prevention, detection, diagnosis, treatment, palliative care and survivorship) and frequency of barriers and facilitators within each health system building-block.

Patient and public involvement

We did not involve patients or the public in the design, conduct or reporting of this study.

Results

Study characteristics

Our search yielded 6384 unique articles, 93 meeting full inclusion criteria (figure 1). There were 67 qualitative (72%), 18 mixed method (19%), 3 quantitative (3%) and 5 informal review studies (5%). Of the five review studies, four were of ‘moderate’ quality and one was of ‘low’ quality. Of the other 88 studies, most were of ‘high’ (n=49; 56%) and ‘moderate’ quality (n=36; 41%), with only two studies being rated as ‘low’ quality (2%). Details of quality assessment can be found in online supplemental appendix C.

Figure 1.

Selection process for literature inclusion.

Articles were published between 1992 and 2020, exhibiting a notable increase in publications after 2005 (figure 2). 73 articles (78%) reported on a single country and 20 (22%) reported on multiple. Our sample represents 86 of 137 LMICs (figure 3).⁶

Figure 2.

Year of publication.

Figure 3.

Geographic distribution of articles.

Notably, 78% (n=73) of articles reported on paediatric NCDs in the ‘other’ category, outside the four principal disease classifications (figure 4). Of these, the most prevalent conditions included metabolic disorders (n=13; 14%), such as inborn errors of metabolism, congenital thyroid disorders and obesity; haematological disorders (n=11; 12%), primarily sickle cell disease and trait, anaemia and haemophilia and otolaryngological conditions (n=11; 12%), principally congenital or childhood-acquired hearing loss. Within the 4 principal classifications, 13 pertained to cancer (14%), 6 to cardiovascular disease (6%), 1 to chronic respiratory disease (1%) and none to diabetes mellitus (0%).

Figure 4.

Non-communicable disease (NCD) classifications. (A) Classification by WHO’s main four NCDs framework; cardiovascular disease (n=6), chronic respiratory disease (n=1), diabetes mellitus (n=0), cancer (n=13) and others (n=73). (B) Classification of NCDs not encompassed by the framework; metabolic (n=13), haematological (n=11), otolaryngological (n=11), child and adolescent mental health (n=8), intellectual development (n=7), musculoskeletal (n=6), ophthalmic (n=6), congenital abnormalities (n=4), nutrition-related (n=3), renal (n=1), neurological (n=1), odonatological (n=1) and rheumatological conditions (n=1).

Most articles reported on a policy or programme addressing a single care continuum component (n=67; 72%), with a high prevalence involving prevention (n=23; 24%), detection and diagnosis (n=47; 51%) and treatment (n=57; 61%) (figure 5). Specific to cancer, palliative care and survivorship were addressed in four and two articles, respectively.

Figure 5.

Components of care continuum represented in studies. NCD, non-communicable disease.

Study characteristics of included articles are detailed in table 1.

Table 1.

Characteristics of included studies

Study	Country	NCD	Care continuum component
Abalo et al 102	Cuba	Hearing loss	Detection/Diagnosis, treatment
Abbo et al 67	Uganda	Nodding syndrome, epilepsy	Treatment
Abu-Osba et al 100	Saudi Arabia	Multiple	Detection/Diagnosis
Akilan et al 28	India	Hearing loss	Detection/Diagnosis
Akintayo et al 63	Nigeria	Clubfoot	Treatment
Akol et al 16	Uganda	Mental health, neuropsychiatric illness	Prevention
Al-Jurayyan et al92	Saudi Arabia	Congenital hypothyroidism	Detection/Diagnosis
Andersen et al 77	Botswana	Eye health	Detection/Diagnosis
Anie et al 42	Ghana	Sickle cell disease/trait	Detection/Diagnosis
Atiim et al 57	Ghana	Food allergy	Detection/Diagnosis, treatment
Barboza-Arguello et al 47	Costa Rica	Neural tube defects	Treatment
Berry et al 45	India	Anaemia	Prevention
Bingoler Pekcici et al 32	Turkey	Developmental disorders	Treatment
Borrajo12	Multiple	Multiple	Detection/Diagnosis
Brewer et al 84	Peru	Anaemia	Treatment
Bussé et al 29	Albania	Hearing loss	Detection/Diagnosis
Calderón-Colmenero et al 99	Mexico	Cardiovascular disease	Treatment
Calderón-Colmenero98	Mexico	Cardiovascular disease	Treatment
Coker-Bolt et al 69	Ethiopia	Physical disability	Treatment
Dearani et al 48	Multiple	Cardiovascular disease	Prevention, detection/diagnosis, treatment
De Garcia et al 70	Multiple	Hearing loss	Detection/Diagnosis, treatment
Denburg et al 93	Multiple	Cancer	Treatment
Denburg et al 10	Multiple	Cancer	Treatment, palliative care, survivorship
Dhillon et al 81	Multiple	Anaemia	Treatment
Diez-Canseco et al 46	Peru	Multiple	Prevention
Dvorak et al 54	Mauritius	Multiple	Prevention
Edefonti et al 40	Nicaragua	Renal disease	Detection/Diagnosis, treatment
Ekeroma et al 49	Multiple	Cancer	Prevention, treatment, palliative care
Ertem et al 87	Turkey	Developmental disorders	Prevention, detection/diagnosis, treatment
Ford-Powell et al 65	Bangladesh	Clubfoot	Treatment
Franz et al 79	South Africa	Autism spectrum disorder	Detection/Diagnosis
Ghodsi et al 80	Iran	Malnutrition	Treatment
Gladstone et al 83	Malawi	Visual impairment	Treatment
Gordillo et al 72	Peru	Retinopathy of prematurity	Prevention
Gregersen et al 11	South Africa	Multiple	Prevention, detection/diagnosis, treatment
Hailu et al 50	Ethiopia	Cancer	Prevention, detection/diagnosis, treatment, palliative care, survivorship
Hamdani et al 51	Pakistan	Developmental disorders	Treatment
Hamdan et al 96	Multiple	Anaemia	Prevention, detection/diagnosis, treatment
Hariharan et al 94	Argentina	Retinopathy of prematurity	Detection/Diagnosis, treatment
Harpham and Tuan60	Vietnam	Mental health	Detection/Diagnosis
Hatam et al 30	Iran	Multiple	Detection/Diagnosis
Hill et al 41	Kenya	Cancer	Treatment
Hobday et al 73	Timor-Leste	Eye health	Prevention
Hockenberry et al 91	Multiple	Cancer	Treatment
Hoehn et al 85	The People’s Democratic Republic of Laos	Multiple	Detection/Diagnosis
Howard et al 97	Multiple	Cancer	Treatment
Khan et al 38	Pakistan	Hearing loss	Treatment
Khapre et al 58	India	Anaemia	Treatment
Khneisser et al 90	Lebanon	Multiple	Detection/Diagnosis
Kleintjes et al 25	Multiple	Mental health	Detection/Diagnosis
Krotoski et al 78	Multiple	Congenital hypothyroidism	Detection/Diagnosis
Kutcher et al 19	Multiple	Mental health	Detection/Diagnosis
Larrandaburu et al 20	Uruguay	Multiple	Detection/Diagnosis
Leong et al 59	Malaysia	Multiple	Detection/Diagnosis
Martin et al 44	Peurto Rico	Chronic respiratory disease	Treatment
McElroy et al 27	Uganda	Clubfoot	Treatment
McKay et al 52	Uganda	Mental health	Treatment
McLean et al 43	Rwanda	Anaemia	Prevention, treatment
Mokitimi et al 86	South Africa	Mental health	Prevention
Moodley et al 101	South Africa	Hearing loss	Detection/Diagnosis, treatment
Mugomeri37	Lesotho	Nutritional deficiency	Detection/Diagnosis
Mumtaz et al 61	Pakistan	Hearing loss	Detection/Diagnosis
Okonta and Tobin-West74	Nigeria	Cardiovascular disease	Treatment
Olusanya and Okolo et al 33	Multiple	Hearing loss	Detection/Diagnosis
Olusanya et al 53	Multiple	Hearing loss	Detection/Diagnosis
Orem and Wabinga et al 56	Uganda	Cancer	Prevention, detection/diagnosis, treatment
Owen et al 68	Multiple	Clubfoot	Treatment
Padilla35	Multiple	Multiple	Detection/Diagnosis
Padilla and Therrell et al 23	Multiple	Multiple	Detection/Diagnosis
Padilla and Therrell et al 62	Multiple	Multiple	Detection/Diagnosis
Park et al 24	The People’s Democratic Republic of Laos	Cancer	Treatment
Pasco et al 89	Romania	Autism spectrum disorder	Detection/Diagnosis, treatment
Penny et al 34	Uganda	Chronic orthopaedic disabilities	Treatment
Picolo et al 82	Mozambique	Nutritional deficiency	Prevention
Pillay and Lockhat et al 55	South Africa	Mental health	Prevention, detection/diagnosis, treatment
Poon and Luke et al 36	China	Haemophilia	Detection/Diagnosis, treatment
Prathanee et al 26	Thailand	Speech and language complications postcleft palate surgery	Treatment
Regmi and Wyber et al 88	Nepal	Cardiovascular disease	Treatment
Roche et al 13	Indonesia	Anaemia	Prevention
Shulman et al 39	Rwanda	Cancer	Detection/Diagnosis, treatment
Shung-King et al 18	South Africa	Multiple	Prevention, detection/diagnosis
Slone et al 75	Botswana	Cancer	Detection/Diagnosis, treatment
Sukati et al 15	Eswatini	Eye health	Prevention, detection/diagnosis, treatment
Sulafa71	Sudan	Cardiovascular disease	Detection/Diagnosis, treatment
Tapela et al 66	Rwanda	Cancer	Detection/Diagnosis, treatment, palliative care
Tekola-Ayele and Rotimi76	Multiple	Multiple	Detection/Diagnosis, treatment
Tekola et al 21	Ethiopia	Developmental disorders	Treatment
Treadwell et al 31	Ghana	Sickle cell disease/trait	Treatment
Vichayanrat et al 14	Thailand	Dental health	Prevention
Whitman et al 64	China	Mental health	Prevention
Wilimas et al 22	Multiple	Cancer	Treatment
Zepeda-Romero and Gilbert et al 17	Mexico	Retinopathy of prematurity	Detection/Diagnosis, treatment
Zhang et al 95	China	Multiple	Treatment

NCD, non-communicable disease.

Implementation factors by health system building-block

Themes in barriers and facilitators are summarised in table 2.

Table 2.

Implementation barriers and facilitators by health system building-block

	Barriers	Facilitators
Health service delivery A health system’s accessibility, continuity and comprehensiveness.7	System organisation and coordination
	System fragmentation Service inaccessibility	Addressing the sociocultural dimensions of healthcare Supplementary measures to promote service accessibility
	Education and communication
	Inadequate public education on paediatric NCDs Limited programme and policy promotion and outreach communication	Thoughtful knowledge translation
	Logistics and infrastructure
	Infrastructural and resource constraints	Integration within existing programmes
Health workforce The human resources engaged in the organisation and delivery of health services.7	Capacity utilisation and optimisation
	High staff turnover and attrition rates Poor staff utilisation	Intersectoral task-shifting Task-shifting to community health workers
	Qualification and training
	Limited domestic training Limited efforts to bolster knowledge retention	Adaptive training delivery
	Support and incentives
	Non-competitive compensation Poorly distributed workloads as a consequence of limited capacities Limited employment opportunity	Diverse incentives
Leadership and governance The policy frameworks; oversight; coalition-building and regulation and accountability processes within a health system.7	Planning and priority setting
	Competing public health priorities Inadequate political visibility	Strategic agenda alignment Policy windows Multiphase implementation
	Policy development
	Insufficient policy specificity	Health ministry support
	Stakeholder relationships
	Poor stakeholder cohesion	Multisectoral coordination Community-level stakeholders Inter-regional and international partnerships
Financing The mobilisation, accumulation and allocation of financial resources to maintain and improve population health coverage.7	Funding streams
	Limited public financing Irregularity of private funding User payment Low-coverage healthcare insurance	Enhancing service affordability
	Financial and economic costs
	High implementation costs Undetermined cost-effectiveness	Investment demand generation
Health information systems Systems that concern the generation, complication and analysis of data that is ultimately synthesised for use and communication.7	Surveillance
	Limited paediatric surveillance data systems Poor data collection capacity	Strong research capability Data sharing
	Data dissemination and management
	Poor data management	Established evaluation and regulatory mechanisms
Access to essential medicines A health system’s procurement process, supply and distribution management and regulatory control of essential therapies, inclusive of medicines, vaccines and technologies.7	Procurement and supply chain management
	Lack of domestic manufacturing capacity Inadequate storage systems	Private-public coordination Proactive procurement processes
	Distribution
	Demand-insensitive distribution	None identified
	Regulation
	Restricted formulary listing of paediatric NCD medicines	None identified
Other Non-health system insights to capture the environmental and societal interactions that influence paediatric health.2	Sociocultural influence
	Inadequate support provided to caregivers Illness stigma	None identified
	Sociopolitical climate
	Political instability Civil insecurity Distrust in healthcare and government systems	Advantageous political-economic windows

Refer to online supplemental appendix D for details and supporting references.

NCD, non-communicable disease.

Health service delivery

Implementation factors related to health service delivery were most frequently reported (figure 6). Shortfalls in service delivery highlighted system fragmentation, service inaccessibility and inadequate public health education. Meanwhile, efforts to mitigate geographic disparities, addressing sociocultural factors and integration within existent services and facilities were key facilitators.

Figure 6.

Frequency of implementation barriers and facilitators across health system building-blocks.

System organisation and accessibility

Care continuity was commonly hindered by system fragmentation across healthcare tiers, and among specialties and service types.^10–26 Authors emphasised the particular relevance of system fragmentation to paediatric NCDs, as children’s development necessitates age-specific services linked with transitioning services. Service inaccessibility was another frequently cited barrier, driven by geographic dispersion of patients from programme sites, routine centralisation of paediatric NCD services and inadequate public transportation infrastructure.18 19 23 27–37 These challenges have outsized bearing on paediatric NCD care, since service delivery is often concentrated at urban health facilities and consistent longitudinal follow-up essential to favourable long-term outcomes is needed.^{15 38 39} Challenges are amplified by childhood dependency on caregivers and the family unit. Studies reported family-level indirect care costs, with caregivers forced to compromise other responsibilities, including income generation, and out-of-pocket costs related to childcare or transportation for other dependents.^{27 29 40}

Key facilitators included addressing sociocultural dimensions of healthcare to improve accessibility.10 13 14 19 21 22 24 24 26 41–54 An Indonesian national iron supplementation programme for adolescent females used formative interviews with them and key actors, such as parents and educators, to enhance the relevance of their awareness campaigns.¹³ A Ugandan rehabilitation programme for paediatric orthopaedic disorders established community-based rehabilitation hostels as stepdown units for rural populations, reducing travel distance and frequency while alleviating healthcare facility congestion.³⁴

Education and communication

Inadequate public health education on NCDs impeded health-seeking behaviours and addressing stigma and misinformation, hindering uptake of implemented strategies.11 12 15 16 19 20 22 36 55–61 Reflecting Ghana’s lack of formalised policy on health promotion, public awareness of food allergy symptoms and risks was limited. Exacerbated by scepticism and misinformation perpetrated by cultural beliefs, paediatric cases were under-reported and addressed outside the formal medical system.⁵⁷ Limited outreach communication to patients, caregivers and healthcare personnel also hindered uptake. Patients and caregivers remained unaware of available services, while healthcare personnel lacked the necessary information to offer recommendations and referrals.^62–64

Logistics and infrastructure

Healthcare infrastructural constraints impeded the development of dedicated paediatric NCD facilities and their integration within facilities already grappling with overwhelming patient traffic and deteriorating sanitary environments.33 63 65–67 Limited or inconsistent supplies in poor and outdated conditions additionally constrained programme service delivery.15 17 22 29 50 57 68–72 Given the limited resources of LMICs, integration with existing paediatric services was advantageous. Nigeria and South Africa integrated newborn hearing screening within immunisation clinics and schedules, leveraging established credibility and enhanced compliance with convenient two-in-one scheduling, while facilitating cost-effectiveness.³³

Health workforce

The deficits in specialty provider capacity limiting paediatric NCD strategies in LMICs were further exacerbated by poor staff delegation and weak sustainability of already limited capacity; obstacles to achieving pediatric-specific education and continued training and the lack of staff support and retention incentives.

Capacity utilisation and optimisation

Many initiatives lacked specialty-qualified or pediatrics-qualified personnel, contributing to healthcare fragmentation.10 15 16 25 26 31 35 51 66 67 70 71 73–79 In the Healthy Eyes in School initiative of Timor-Leste’s Alieu district, the presence of only one eye care technician from 2011 to 2014 created a referral bottleneck, limiting large-scale implementation.⁷³ Systems with sufficient capacities were often compromised by poor staff retention or frequent rotation, perpetuating a cyclic need for recruitment and specialty training unconducive to programme sustainability.13 15 32 50 58 80 Poor staff allocation also resulted in inefficient use of skills; a paediatric cardiology programme in Sudan failed to use available echocardiogram technicians, sustaining dependence on physicians to conduct diagnostic testing.⁷¹ Task-shifting extended beyond the health sector was a facilitator in various community-based strategies.28 33 39 43 55 65 66 81 Non-medical tasks, such as information dissemination, were purposefully delegated to teachers and childcare personnel in a community-based hearing screening project implemented in the Tamil Nadu district of South India.²⁸ Leveraging the community presence and familiarity of these workers facilitated acceptance and compliance while alleviating the burden on medical personnel. Community health workers (CHWs) were a valuable resource for task-shifting. Embedded in the community and having clinical training, they mitigated concerns about qualifications that often impede the acceptability of task-shifting.34 43 51 52 59 81–83

Qualification and training

Capacity limitations due to barriers in qualification and training were pronounced for paediatrics. The lack of domestic medical and specialty education compelled prospective trainees to pursue paediatrics programmes and employment in neighbouring regions or foreign countries.^{15 38 49} Furthermore, implementing innovations was often constrained by the limited number of staff who were supplementally trained to use new strategies in paediatric care.10 12 16 18–20 39 59 63 84 Uruguay’s development of comprehensive and universal care policies for congenital and rare NCDs highlighted the critical need to improve domestic health personnel training in the identification of congenital anomalies.²⁰ Expert-knowledge gaps exacerbated frailties along the care continuum, as treatment-focused initiatives were constrained by limited capacity for detection and diagnosis of paediatric NCDs. Insufficient attention to knowledge retention, evidenced by lack of refresher training and feedback opportunities, created an additional barrier.12 14 18 24 26 29 31 57 58 67 77 83 85–88 Specific to community task-shifting, retention of teachers and CHWs was compromised as they were expected to implement novel responsibilities without standardised guidance and instruction.⁵⁸

Significantly, adaptive and flexible delivery of training was a powerful facilitator of paediatric NCD programmes.11 13 14 16 24 26 31 34 41 48 50 71 72 75 82 87 89–92 Versatile platforms, such as tele-teaching or multimethod training (eg, onsite and offsite, and sessions by request), enhanced the feasibility of continuous training and broadened available training options.

Workforce support and incentives

Recruitment was limited by uncompetitive public sector salaries, which were routinely well below domestic and foreign private sector remuneration.15 17 24 32 49 50 77 Private-public compensation disparities resulted in limited recruitment and retention of childhood cancer providers in the Philippines’ public health sector.⁹³ Additionally, reliance on volunteerism or low wages reduced workforce retention.^{32 51 94} Increased burden on the public paediatric workforce was identified as a downstream barrier to NCD programme implementation in a range of health systems. Staff shortages placed excessive burdens on a limited few, resulting in high patient-to-nurse ratios, which led to increased risk of medical errors and workforce burnout.15 18 46 56 58 59 63 67 73 80 91 Persistent lack of student interest in paediatric specialty training schemes, based on limited employment opportunities due to domestic underdevelopment of these specialties, was also an important barrier.^{26 50} This perpetuated a vicious cycle hindering the growth of paediatrics.

Leadership and governance

Health system leadership and governance emerged as decisive determinants of paediatric NCD policy implementation, underscoring the importance of political recognition and priority, paediatric policy specificity rather than umbrella NCD policies and the inclusion of a broad panoply of stakeholders in the implementation of policies and programmes.

Planning and priority setting

Successful implementation of paediatric NCD policies and programmes relies on sustained prioritisation within national health system planning cycles. In many LMICs undergoing epidemiological transition, NCD prioritisation is challenged with a persistent focus on communicable diseases.10–12 15 24 35 53 59 60 62 82 A key barrier was inadequate political visibility of paediatric NCDs, evidenced by lack of awareness among government officials regarding disease burden, of pediatric-focused policies and of supportive policies to facilitate the integration of paediatric programmes in health systems.10 14 15 17 19 24 26 32 35 42 49 50 56 61 62 67 68 74 75 78 79 82 83 89 95

Conversely, strategic framing of policies to align with current policy agendas facilitated prioritisation: they met less resistance, drew on existing momentum and leveraged stakeholders already invested in aligned policies and programmes.18 24 35 39 47 50 52 60 64 79 Policy windows were powerful facilitators of strategic alignment. A school-based programme promoting mental health among Chinese students was implemented in tandem with overarching government efforts to reform its education system.⁶⁴ Finally, multiphase implementation of paediatric NCD policies, through pilot initiatives and evidence-based reporting of sequential phases, surfaced as a facilitator, enabling the timely identification of barriers, informing subsequent planning and offering increased opportunities for stakeholder input.43 44 54 69

Policy development

Lack of specificity in setting objectives was a barrier in programme and policy development.15 76 79 80 96 In 2005, Iran implemented a national programme to enhance paediatric nutrition in impoverished settings.⁸⁰ Despite having a clear mission statement, it failed to implement successfully due to the absence of actionable protocols and to ambiguity in participant selection criteria, family socioeconomic status assessment and nutrition recovery indicators. This resulted in unwarranted exclusion of children, underscoring the necessity of designing paediatric-specific strategies rather than adopting pre-existing umbrella or adult frameworks and policies.

Stakeholder relationships

Health-promoting school programmes uniquely highlighted the multisectoral coordination required between the education and healthcare sectors to ensure feasible curriculum integration, appropriate training and sustainable allocation of responsibility.18 19 25 58 64 Local stakeholders often played a crucial role in sustainable implementation.20 21 42–44 48–50 52 60 65 69 81 94 97 The My Child Matters programme by the Sanofi Espoir Foundation recognised the synergistic and catalytic role of local non-governmental organisations that leverage pre-existing community acceptance and resources and that sustain domestic leadership, mitigating chronic dependence on external stakeholders.⁹⁷ Moreover, community leaders, including religious and elder figures, facilitate culturally sensitive implementation, leveraging their heightened influence in LMICs and on paediatric social development.^{21 42 44} Inter-regional and international partnerships fostered policy community cohesion, in turn facilitating political advocacy and the sharing of high-level information and resources.10 23 24 32 36 38–41 48 49 71 75 76 78 90 91 93 97 Twinning partnerships with HICs were a prevalent approach to enhance paediatric NCD programmes in LMICs, facilitating increased awareness and mobilisation of supplemental resources and capacity.23 38–40 49 71 90 91 93

Financing

The chronicity and sequelae of paediatric NCDs highlights necessary but commonly unavailable funding for sustained programme implementation, including across developmental stages, in LMICs. A heavier family financial burden also exists, given recurrent user payments, larger family units and limited public insurance schemes. In health systems where private services are prevalent, provider concerns of non-profitability and reduced service uptake exist.

Funding streams

Limited public financing of paediatric NCD initiatives was frequently attributed to low political visibility and priority.10 13 16 22 38 50 53 59 61 63 78 80 82 93 95 Consequently, supplementary private funding, often from HICs and donor organisations, was commonly sought. However, funding flow was often inconsistent or limited to short-term agreements.13 16 22–24 40 41 43 44 53 54 68 97 98 At the clinical interface, out-of-pocket healthcare expenditure exacerbated inaccessibility in health systems reliant on user fees, especially for larger families.27 36 38 74 85 The lack of or limitations in coverage of public health insurance perpetuated this burden. Low-coverage schemes typically had limited inclusion of NCDs or specialised treatments, restricted availability to certain populations or had wide coverage but with low compensation.29 36 39 70 71 87 93 95 To this end, the Shanghai Children Hospital Care Aid was implemented to address the reluctance of commercial insurance companies and charitable organisations in China to cover costly treatments for critical paediatric diseases, including NCDs.⁹⁵

Innovative strategies to enhance service affordability facilitated population accessibility, and programme and policy uptake.17 20 23 39 41 48 66 81 One example is the microfinance programme developed by the National Health Insurance Fund under Kenya’s national retinoblastoma strategy to improve paediatric outcomes. Caregivers of inpatients were taught to handmake and sell crafts, the profits of which were used to cover their health insurance premiums and additional treatment costs.⁴¹

Financial and economic costs

High costs, such as expensive upfront capital investments in diagnostic technologies, were a barrier for initial investments, profitability and sustainability of programmes.33 59 71 95 Newborn screening programmes in Latin America struggled to source screening tests as private suppliers were often concerned about poor profitability.¹² Furthermore, the failure to determine programme cost-effectiveness deterred appropriate evaluation and design of pre-emptive strategies to decrease costs.^{12 20 39} Whether through the establishment of efficiency through cost-utility analyses or demonstration of profitability to stakeholders, dedicated efforts to generate investment demand was a financial facilitator to implementation.^{13 23 30}

Health information systems

As a cross-cutting building-block, the underdevelopment of health information systems, demonstrated by poor surveillance and data management systems, contributed to political invisibility of paediatric NCDs and impeded effective service delivery.

Surveillance

The lack of established data systems for surveillance, through registries or epidemiological modelling, hampered political visibility and prioritisation of paediatric NCDs and was a contributing barrier to programme planning and delivery.11 20 39 47 49 60 61 75 81 82 93 98–100 A policy assessment of Pakistan’s public neonatal hearing screening determined a deficit of credible research on paediatric hearing impairment that limited policymaker awareness of the associated health and economic burdens.⁶¹ Limited data collection capability was a barrier to addressing such information gaps.10 13 16 24 37 53 57 62 70 81 82 99–101 Limitations were technical and infrastructural, for example, the absence of centralised data management systems, as well as resource-based, including inadequate human resource capacity or funding.10 55 78 80 101

Strong research capability served as a facilitator, mutually reinforcing surveillance efforts.39 41 56 67 This is exemplified by Uganda’s rich cancer research history, from which paediatric epidemiology justified and informed the incorporation of paediatric cancer priorities in the national cancer control programme.⁵⁶ Data sharing, enabled by collaborative data-pooling networks and open information exchange among paediatric clinical and research groups, facilitated expansion of epidemiological information for LMICs struggling to develop domestic health information systems.13 40 44 48 56 62 98

Data dissemination and management

Existing information systems for paediatric NCDs were commonly compromised by poor management, evinced by inconsistent data completeness and accuracy, data loss during transfer and delayed error responses.15 19 29 43 47 65 66 80 99 Costa Rica’s implementation of its registry for congenital anomalies was challenged by participating hospitals submitting case reports in inconsistent formats, without verification and at erratic intervals.⁴⁷ Conversely, pairing NCD information system development with mechanisms for oversight, and routine monitoring and evaluation facilitated quality assurance of data systems.48 60 68 73 76 81 89

Access to essential medicines

The included literature highlights a complex interplay of determinants impacting access to medicines for paediatric NCDs. Principal barriers include gaps in medicine availability and price variability conditioned by reactive procurement processes, reliance on foreign suppliers, weak supply management systems, poorly coordinated distribution and limited listing of paediatric NCD drug indications and formulations on national formularies.

Procurement and supply chain management

Paediatric NCD medicine availability across an array of diseases and systems was characterised by recurrent stockouts, and unreliable supply.10 15 27 38–40 56 58 67 71 75 80 81 88 Two dominant barriers were identified: the lack of domestic manufacturers and resulting reliance on importation, in which lengthy procurement times compromised quality and sustained availability; and inadequate storage systems that precipitated preventable product expiry and spoilage.43 45 81 88 By contrast, coordinated procurement strategies for private and public sourcing leveraged policy-backed resource allocation and lowered financial barriers through economies of scale.13 22 41 65 66 81 Proactive procurement processes, including those premised on evidence-based forecasting, mitigated shortages and helped stabilise prices in the context of relatively low demand for paediatric NCD drug indications.13 16 65 67 102

Distribution

The distribution of NCD medicines to paediatric facilities or programmes was sometimes compromised by inaccurate forecasting, discordance between supply duration and restocking intervals, insufficient allocation and non-standardised distribution.14 16 43 45 58 67 82 A school-based supplementation programme in India revealed that all participating schools had experienced stockout periods or expired products as distribution scheduling was misaligned with the school year, underscoring multisectoral coordination as a key facilitator of policy implementation.⁵⁸

Regulation

Limited inclusion of paediatric indications and formulations of NCD medicines on national formularies and essential medication lists hampered reliable public sector access for children.^{10 16} A stakeholder forum on implementation considerations for national childhood cancer strategies in Latin America highlighted regulatory policies that constrained formulary inclusion of certain medicine formulations and devices, in the absence of overarching policy prioritisation of childhood cancers in national cancer control plans.¹⁰ Health technology regulation also created financial barriers to access, demonstrated by the mark-ups and taxes imposed on medical supplies by Ethiopia’s governmental agencies, which complicated supplying materials essential for the implementation of a paediatric physical and occupational therapy training programme.⁶⁹

Other factors

Paediatric health and healthcare are heavily influenced by the environments and communities that children and youth interact with in their everyday lives. Sociocultural values, political instability and macroeconomic conditions were encompassing determinants of paediatric NCD policy and programme implementation.

Sociocultural influence

Inadequate caregiver support was a sociocultural barrier that hindered uptake and compliance with implemented strategies.21 27 31 34 80 84 Caregivers faced unshared financial burdens, challenges in acquiring childcare, overwhelming or competing parental responsibilities and misleading social interactions that discouraged health proactivity.21 27 34 63 84 The latter was heavily influenced by societal stigma.19 21 26 31 33 38 51 55 63 67 67 83 86

Sociopolitical climate

Political instability compromised health resources and services.23 55 59 71 80 90 Iran’s frequent administration turnovers with each presidential election introduced numerous political priority shifts that jeopardised the stability of paediatric NCD policies and programmes.⁸⁰ Civil insecurities, such as risk of imminent war and electrical outages, also obstructed the implementation and accessibility of essential health services.^{55 90} Institutional distrust also surfaced as a barrier, as the uptake of some healthcare programmes was tainted by increasing prevalence of child sexual abuse or hindered by societal distrust in governments.31 35 55 59 67 84 Conversely, in the development of Shanghai’s national child health security system for critical diseases, analysis of existent insurance schemes identified periods of national stability and rapid economic development as advantageous economic windows for increased financing.⁹⁵

Discussion

We conducted a systematic review of paediatric NCD policies and programmes in LMICs to identify factors that promote or forestall their successful implementation. In the 93 studies, a wide range of paediatric conditions was represented, but disproportionately outside of the conventional adult NCD burden, highlighting the epidemiological differences and unique attendant health system considerations relevant to children and youth with NCDs. Synthesising our findings across the six WHO health-system building-blocks, we identified facilitators instrumental for policy and programme implementation, prominent among them appropriate consideration of sociocultural influences, comprehensive linking of care components, strategic alignment with broader policy agendas and multisectoral coordination, particularly with the education sector. Conversely, health system fragmentation—across tiers of care and between service settings—and lack of dedicated resource allocation to paediatric NCD management were common barriers, conditioned by weak evidence of paediatric NCD burden, corollary lack of sociopolitical visibility and resultant failure of political prioritisation. Notably, prominent barriers and facilitators across all health system domains highlight the unique diversity of stakeholders in paediatric NCD care, emphasising the influence of family, community and education systems on healthcare access for children and youth.

Previous reviews of NCDs in LMICs have focused on adults; in those that included children, we have not found any related to policy and programme implementation. A review by Jain et al examined a broad range of NCD interventions for school-aged children to assess their clinical but not implementation, effectiveness and only 12.5% of included studies were LMIC-based.¹⁰³ Employing a health-system building-blocks framework, Heller et al reviewed the delivery of NCD care by non-physician health workers in LMICs, a programmatic facilitator identified in our review; however, their focus was not on children.¹⁰⁴ Shah et al reviewed the delivery of NCD care to women and children in conflict zones. Although the included studies were based in LMICs, many of the insights identified were unique to conflict settings, limiting the relevance of their findings to policy and programme implementation for paediatric NCDs in LMIC health systems more generally.¹⁰⁵

In both our study and in the general literature on paediatric NCDs in LMICs, we note an important gap in programmes and policies to address diabetes, a growing burden worldwide. This may relate to limited study of the incidence of type 1 diabetes in LMICs, particularly sub-Saharan Africa, resulting in lack of evidence for policy guidance.¹⁰⁶

Our review underscores the relative neglect of paediatric NCDs in global NCD policy. The ‘4×4’ framework (four main NCDs of cardiovascular disease, diabetes, chronic respiratory diseases and cancers and four main modifiable risk factors of tobacco use, excessive alcohol use, unhealthy diets and physical inactivity) and its ‘5×5’ extension (including mental health and air pollution) have excluded diseases and issues pertinent to LMICs.¹⁰⁷ The pervasiveness of this framing in policy agendas has poorly served children and youth in LMICs by relegating the majority of paediatric NCDs to a poorly specified category of ‘other’. It also ignores the fact that most paediatric NCDs are driven not by the causal and modifiable risk factors at play in adult NCDs, but by genetic, congenital and poorly understood environmental factors less amenable to behavioural risk reduction or prevention. The Lancet NCDI Poverty Commission estimated that the NCD DALY loss per 1000 persons among those aged under 5 years in LMICs is greater than for the same age group in HICs and similar in those aged 5–19 years, and has recommended disaggregation of the maternal and child mortality targets to better address NCDs across health system contexts.¹⁰⁸

The epidemiology, presentation and management of paediatric NCDs are distinct from those in adults and often age-specific, necessitating unique modes and patterns of engagement with healthcare systems. Different frameworks for paediatric NCD policy and programme implementation and evaluation are therefore required.² The Paediatric Oncology System Integration Tool framework delineates the unique realities and programmatic requirements of childhood cancer care in a system-wide perspective, assessing cancer system performance through integrated consideration of the health workforce, infrastructure, medicines and technologies and supportive services.¹⁰⁹ Comparable heuristics for other paediatric NCDs will help develop and evaluate evidence-informed policies and programmes for children and youth in LMICs.

The prevalence of school-based programming and policies in our review highlights the education sector as a critical health system extension for the paediatric population.¹¹⁰ However, in the least-developed countries, approximately 81% of children do not attend early childhood education.¹¹⁰ Students in LMICs are also disproportionately impacted by civil insecurity, as demonstrated by the average of 4 months of disrupted schooling during the initial year of the COVID-19 pandemic, in contrast to the average 6 weeks lost in HICs.¹¹¹ Minimising attendance disruptions may have far-reaching impacts on health policy and programming.

School-based and community-based approaches in particular have contributed to successful child and adolescent mental health (CAMH) programming.² This underscores mental health as a key multisectoral investment priority, given that CAMH is the most prevalent NCD among those under the age of 20.² A UNICEF analysis reported that school-based interventions addressing anxiety, depression and suicide provide a return on investment of US$21.5 for every US$1 invested over 80 years.¹¹⁰ Yet, our review revealed only eight CAMH-focused studies, pointing to stigma, lack of epidemiological evidence and a scarcity of mental health researchers and providers in LMICs.¹¹⁰ The need to increase visibility for mental health underscores leadership and governance as a cross-cutting implementation factor, as invisibility hinders policy prioritisation and resource allocation.¹⁰⁹

Strengths

This is, to our knowledge, the first systematic review of the factors contributing to the success of paediatric NCD policies and programmes in LMICs. It provides much-needed evidence to inform a life-course approach to addressing the growing strain of NCDs in low-resource settings. With its wide geographic span, our review reports on strategies that provide a representative view of implementation considerations in low-resource health systems. Our study benefited from a structured system-wide lens, particularly needed to capture the intersectoral determinants of NCDs and the wider ecosystem of NCD healthcare particular to children and youth.

Limitations

Our review is limited by the inclusion of English-only articles. We may have missed country-specific data contained in locally published, non-English journals. We also acknowledge the potential loss of valuable proximal insights of LMIC-based collaborators. However, our review includes studies conducted by authors and teams based in LMICs, ensuring local perspectives are incorporated. Moreover, in excluding literature on injuries, we may have missed emerging and pertinent information.

The WHO building-blocks framework offers a useful skeleton with which to map complex health systems; however, implementation challenges rarely manifest themselves in clear-cut blocks. Consequently, classification of barriers and facilitators was subjective, and their categorisation may have varied for articles where settings were not explicit. Inter-related or cross-cutting themes may also have been constrained by this framework. However, we deliberately identified boundary interactions across blocks to overcome overcompartmentalisation.

Conclusion

The implementation of NCD policies and programmes for children and youth face a multitude of barriers in LMICs across all aspects of the health sector. Yet, key facilitators identified offer an opportunity to advance our understanding of effective NCD policy and programme implementation among paediatric populations. Future implementation research focused on specific policies and programmes is recommended to further discern which factors have the most consequential impact within differing local contexts. Moreover, given the paediatric NCD burden in LMICs, future studies may strive to improve research coverage for countries not included in the current study and for understudied impacts on childhood development.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information. Full extracted datasets are available on reasonable request.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Not applicable.