Interventions for High-Burden Infectious Diseases in Children and Adolescents: A Meta-analysis

The authors of this study aim to synthesize existing evidence on interventions for high-burden infectious diseases among school-aged children and adolescents aged 5 to 19 years.

This systematic review follows the guidelines recommended by the Preferred Reporting Items for Systematic Reviews and Meta-analyses.¹¹  The Preferred Reporting Items for Systematic Reviews and Meta-analyses checklist is presented in Supplemental Table 1.

We included randomized controlled trials (RCT) (both cluster- and individual-level randomization), quasi-experimental studies, and controlled before-after studies assessing interventions for improved detection and/or case management of high-burden infectious diseases. Our target populations were school-aged children and adolescents aged 5 to 19 years from across the globe. Studies including participants with a mean or median age falling in our age group of interest were included, as were studies that reported outcomes according to age categories. We included the following high-burden infectious diseases based on the Global Burden of Disease Study 2019 for this age group: diarrhea, cholera, shigellosis, Campylobacter, acute respiratory infections (ARI), adenovirus, Hib, pertussis, typhoid or paratyphoid intestinal infections, malaria, HIV and AIDS, tuberculosis, measles, meningitis, syphilis, hepatitis B and C, and UTIs. Any intervention targeted toward prompt detection or effective management of these diseases were included. Examples of interventions of interest include counseling and awareness, water treatment, shorter course of treatment of better compliance, prophylactic treatments for disease prevention, micronutrient supplementation as adjunct therapy, and improving detection and management. We included studies that were conducted after the year 2000 to focus on the current plausible management strategies.

We excluded studies that targeted noninfectious causes of morbidity and mortality in this age group (such as neoplasms, road traffic accidents, etc), or were addressing interventions only toward upper respiratory tract infections (eg, otitis media, tonsillitis, pharyngitis, etc) or vaccinations. We also excluded studies assessing drug safety and efficacy.

We included studies that satisfied our eligibility criteria and reported outcomes on the selected high-burden diseases. Studies that reported the effects of the intervention on detection or management of the selected diseases, including detection rates, compliance to medication, duration of illness, need for hospitalization, incidence and prevalence of high-burden diseases, time is taken for resolution of symptoms, etc, were included. Studies reporting only on behavior change outcomes, such as frequency of handwashing etc, were excluded from the review.

We conducted an electronic search until December 31, 2020, using PubMed, Cochrane CENTRAL, and Google Scholar and exported the search results to the Covidence application.¹²  A few of the terms used in the search strategy were as follows: “case management,” “infectious disease,” “school-age children,” “adolescent,” etc. The complete search strategy is presented in Supplemental Table 2. We also screened the reference lists of relevant reviews to identify any additional studies that might have been missed in the electronic searches.

Two reviewers (D.S.K. and R.N.) screened titles and abstracts on Covidence.¹²  We retrieved the full text of all remaining studies and screened them based on the eligibility criteria. Any conflicts were resolved by reviewers R.A.S. and J.K.D. Data were extracted from the included studies on study background, population and setting, intervention group details, comparison group details, outcomes, and other relevant information. We attempted to conduct a meta-analysis (where possible) for dichotomous and continuous variables using Review Manager version 5.1.4.¹³  For continuous data, we used mean difference and for dichotomous data, we used risk ratios (RR) with a 95% confidence interval (CI). Heterogeneity was assessed using I2 and τ2 values, along with visual inspection of forest plots. We used the Cochrane risk of bias tool for the quality assessment of the included RCTs.¹⁴  This tool assesses selection bias, performance bias, detection bias, attrition bias, and reporting bias. Every study was rated to be at “high,” “low,” or “unclear” risk of bias for each component of the risk of bias tool. We summarized the quality of evidence that we extracted using the Grading of Recommendations, Assessment, Development, and Evaluation criteria.

Our search yielded a total of 15 320 results that underwent title and abstract screening. A total of 1550 studies were selected for full-text review, and after cross-referencing of included articles and relevant systematic reviews, we included a total of 31 studies as shown in Fig 1.

A total of 31 studies were included, of which 20 were individually RCTs, 10 were cluster RCTs, and 1 study was a crossover trial.^15–45  All studies were conducted between the years 2002 and 2020. The studies were conducted in the high-, middle-, and low-income countries. Among the included studies, 4 studies were conducted in India, 3 from Mongolia, 3 from Kenya, 2 each from Bangladesh and Pakistan, 1 study each from Colombia, Cambodia, Congo, Egypt, Ethiopia, Greece, Guatemala, Guinea-Bissau, Iran, Peru, Philippines, Tanzania, Uganda, and United States; whereas 3 studies were multicounty studies (1 conducted in Bolivia and Bosnia and 2 studies from Uganda and Zimbabwe). Eighteen studies were conducted in a community setting, 5 in schools, 4 in outpatient settings, and 4 in inpatient settings. Out of the 31 included studies, 14 studies targeted individual participants, whereas the remaining 17 targeted whole communities (including households, schools, diagnostic centers, etc.) but reported the outcomes separately for our population of interest.

The included studies in this review targeted 81 596 individuals; with 1 single study including 44 451 individuals.⁴²  Five studies failed to mention the number of individuals and reported only the number of households included in each study. In terms of diseases targeted; a total of 16 studies targeted diarrhea,^{17,18,20–26,28–30,37,43–45}  6 targeted tuberculosis,^{15,19,31–33,39}  2 targeted HIV,^34,35  2 targeted measles,^27,36  1 study each targeted ARI,¹⁶  malaria,³⁸  and UTI,⁴⁰  whereas 2 studies targeted multiple diseases.^41,42  We did not find any study targeting any other high burden infectious diseases among this age group. The interventions assessed in these studies were WASH interventions (n = 15), micronutrient supplementation (n = 9), antibiotic prophylaxis or shortened course of antibiotic treatment (n = 4), and home or community-based managements (n = 3). Among the studies assessing WASH interventions, the interventions included filter use (n = 5),^{18,29,43–45}  hand hygiene (n = 5),^{21,24,25,30,42}  building water and sanitation facilities (n = 1),³⁷  and water disinfectant use (n = 4).^17,20,22,28  Among studies assessing micronutrient supplementation, 3 studies assessed supplementation with vitamin D3 for tuberculosis and ARI,^15,16,39  5 studies assessed zinc supplementation, and 1 study assessed multiple micronutrients along with zinc supplementation for pneumonia and diarrhea.¹⁹  Home or community-based management strategies included counseling and contingency contract (incentives for children adhering to prescribed treatment) (n = 1)³¹  and increasing awareness among health care workers (n = 2).^32,38  Among studies assessing use of antibiotics, 1 study assessed the impact of shortening the antibiotic course,³³  whereas 3 studies assessed the use of cotrimoxazole as prophylaxis in HIV patients (n = 2)^34,35  and prevention of measles complications (n = 1).²⁷  The table for the characteristics of included studies is added as Supplemental Table 3. Figure 2 summarizes the risk of bias of the included studies.

We report our findings according to the diseases. We could not conduct a meta-analysis for most of the interventions evaluated under each disease’s category (except diarrhea) because of the variety of the interventions included and the outcomes reported.

A total of 16 studies were included assessing the interventions for diarrhea among school-aged children and adolescents age 5 to 19 years.^{17,18,20–26,28–30,37,43–45}  Fourteen studies assessed the impact of WASH interventions on the incidence of diarrhea, whereas 2 assessed the use of zinc supplementation on diarrhea duration.^23,26  Among studies assessing WASH intervention, 5 studies evaluated the impact of different water filtration techniques on diarrhea incidence,^{18,29,43–45}  4 studies used water treatment with disinfectants as an intervention,^17,20,22,28  1 study assessed the impact of building sanitation infrastructure and the provision of hygiene kits on diarrhea-related school absence,³⁷  and another 4 assessed impact of hand hygiene practices.^21,24,25,30 

Findings suggest that water treatment probably reduces diarrhea by 39% (RR, 0.61; 95% CI, 0.49–0.75; 4 studies; I2 0%; moderate evidence quality; Fig 3). The different types of water treatments used are highlighted in Supplemental Table 3; these included lifestraw filter treated water, ceramic water purifier, iron-rich ceramic purifier, and water treatment with concrete BioSand filter. We are uncertain the effect of hand hygiene on diarrhea (RR, 0.79; 95% CI, 0.61–1.03; 1 study; very low-quality evidence; Fig 3).

All of the studies assessing the effect of water filtration techniques reported a reduction in diarrheal illness,^{18,29,37,43,45}  except 1, which reported inconclusive findings due to limited evidence.⁴⁴  The reduction in longitudinal prevalence ratio ranged between 34% and 52% in school-aged children and adolescents aged 5 to 15 years. For zinc supplementation, 1 study supplementing 30 mg of zinc reported a significant reduction in duration of diarrhea (64.1 hours in zinc supplementation group versus 72.8 hours in the control group, P = .028),²⁶  whereas the other study supplementing 20 mg zinc reported no reduction in episodes of diarrhea or its duration (hazard ratio: 0.89; 95% CI 0.63–1.24).²³  Building proper water and sanitation infrastructure in schools led to a significant reduction in odds of school absence due to diarrhea (OR = 0.10; 95% CI, 0.05–0.22; P < .001).³⁷ 

We identified 6 studies assessing interventions for tuberculosis.^{15,19,31–33,39}  Two studies assessed the effect of vitamin D3 supplementation,^15,39  2 studies assessed improvement in compliance of antituberculosis treatment^31,33  (1 used counseling strategies as an intervention,³¹  whereas the other assessed shorter antituberculosis treatment durations and its impact on compliance),³³  1 study assessed the effect of multiple micronutrients and zinc supplementation,¹⁹  and 1 study assessed the effect of increasing awareness and training health care workers on tuberculosis detection rates.³² 

Vitamin D3 supplementation did not lower the risk of tuberculosis infection (RR, 1.10; 95% CI, 0.87–1.38; P = .42) or tuberculosis disease (RR, 0.87; 95% CI, 0.49–1.55) when compared with placebo, and the incidence of adverse events did not differ between the 2 groups.¹⁵  Vitamin D3 supplementation significantly improved growth in stature with mean increase in stature of 2.9 ± 1.6 cm in the vitamin D group and 2.0 ± 1.7 cm in the placebo group (95% CI, 2.16–2.81; P = .003), along with fewer tuberculin skin test conversions from negative to positive (RR, 0.41; 95% CI, 0.16–1.09; P = .06).³⁹  Multiple micronutrient supplementation, zinc supplementation, or multiple micronutrients, including zinc supplementation did not have an impact on anthropometric indices and chest radiograph improvement compared with placebo. However, children who received micronutrients had a faster gain in height over 6 months of intervention compared with those who did not receive micronutrients (change in height-for-age z-score = 0.08; P = .014).¹⁹ 

Peer counseling along with contingency contract (84.8%) and peer-counseling alone (80.3%) led to a significantly higher treatment completion rate, compared with usual care (77.8%).³¹  Shorter course combination therapy (4 months compared with 9 months) led to increased compliance (P = .011 for 4 months versus 9 months course). The radiographic findings of participants from the standard care group were also more commonly suggestive of possible active disease as compared with those from shorter course regimens (P = .001 for 9 months versus 4 months).³³  Increasing general awareness and training of health care workers regarding tuberculosis detection reported a 3-fold increase in tuberculosis detection in the intervention group from 3.8% to 12% when compared with its baseline.³² 

We included 2 studies assessing interventions targeting HIV in this age group.^34,35  Both studies assessed the continuation or cessation of cotrimoxazole antibiotics among children as part of the antiretroviral research for Watoto trial. tuberculosis risk was reported to be higher among children who stopped cotrimoxazole therapy after 96 weeks of commencing antiretroviral therapy (ART) compared with those children who continued therapy (hazard ratio [HR]: 3.0; 95% CI, 1.1–8.3; P = .028).³⁴  Continuation of cotrimoxazole therapy after 96 weeks of ART was also found to be beneficial in reducing hospitalizations for malaria or infections not related to malaria compared with those who stopped cotrimoxazole prophylaxis (HR,1.64; 95% CI, 1.14–2.37; P = .007).³⁵ 

We included 2 studies assessing the interventions for the prevention of complications due to measles, particularly pneumonia in children with measles.^27,36  One study evaluated the effect of zinc supplementation on the duration of measles rash and the time taken for resolution of symptoms³⁶ ; while the other study assessed the effect of cotrimoxazole in children with measles on the development of pneumonia or other severe symptoms and need for hospital admission.²⁷ 

Zinc supplementation with 20 mg tablets taken once daily for 6 days did not provide additional benefit in the treatment of children with measles accompanied by pneumonia as compared with standard treatment with antibiotics and vitamin A therapy alone.³⁶  In children with measles, prophylactic treatment with cotrimoxazole was observed to prevent measles complications, with fewer lesser pneumonia cases (odds ratio, 0.08; 95% CI, 0–0.56).²⁷ 

We included only 1 study that evaluated the impact of vitamin D3 fortified milk on ARI.¹⁶  Findings from this study suggest that milk fortified with vitamin D3 significantly reduced the risk of ARIs among children (RR, 0.52; 95% CI, 0.31–0.89).¹⁶ 

We included 1 study assessing the use of rapid diagnostic tests (RDT) by community health workers (CHWs) and the number of children prescribed artemisinin-based combination therapy (ACT). CHWs were trained to give out ACT based on RDT results or clinical diagnosis alone, suggesting that RDT use by CHWs can be safe and effective for targeting ACT treatment in patients with uncomplicated malaria. This also led to a significant reduction in ACT prescription (by 45%) compared with the group which was treated based on clinical diagnosis alone.³⁸ 

We included a single study that was conducted on hospitalized children with UTI,⁴⁰  assessing the effectiveness of zinc supplementation in treating children with UTI. It was reported that once daily 1 mg/kg per day of zinc supplementation in addition to standard treatment led to a faster recovery with a positive effect in reducing symptoms such as dysuria and urine frequency. However, zinc supplementation was also noted to exacerbate abdominal pain in children and increased its duration.⁴⁰ 

Two studies assessed the impact of interventions on 2 or more high-burden infectious diseases.^41,42  One study assessed the impact of zinc and iron supplementation on the incidence of malaria, ARI, and diarrhea.⁴¹  The other study assessed the impact of hand hygiene on the incidence of influenza-like-illness (ILI), school absenteeism due to ILI, diarrhea, and conjunctivitis.⁴²  Findings from these 2 studies suggest that supplementation with 20 mg zinc reduces diarrhea morbidity by 23%, while iron supplementation was reported to increase morbidity due to malaria and diarrhea in children. There was no effect of micronutrient supplementation on ARI or anthropometric indices.⁴¹  Hand hygiene led to a 50% reduction in laboratory-confirmed influenza disease (P < .0001), a 40% reduction in school absences due to ILI, a 30% reduction in absences due to diarrhea, and 67% reduction in conjunctivitis, when compared with the control group (P < .0001).⁴² 

Our review summarizes evidence on interventions to improve detection and/or effective case management for high-burden infectious diseases among school-aged children and adolescents aged 5 to 19 years. We included a total of 31 studies including 81 596 participants and the interventions assessed in these studies were WASH interventions, micronutrient supplementation, antibiotic prophylaxis or shortened course of antibiotic treatment, and home or community-based management. All of the included studies were judged to be at low risk of bias for sequence generation, although a majority of the studies were judged to be at high or unclear risk of bias for allocation concealment and blinding of participants or personnel and outcome assessors. Most of the studies were judged to be at unclear risk of bias for attrition due to the lack of reported data.

We could not perform meta-analysis for a majority of the interventions evaluated under each disease category (except diarrhea) because of the variances in the interventions included and outcomes reported. Our review findings suggest that for diarrhea, water treatment probably reduces diarrhea by 39%, whereas the effect of hand hygiene on diarrhea is uncertain. Water filtration techniques reduce diarrheal illness by 34% to 52%. Building water and sanitation infrastructure, including water storage systems, latrines, hand-washing facilities, and water points indicated that the odds of being absent from school with diarrhea was almost 10-fold lower. Supplementation with 30 mg of zinc led to a significant reduction in the duration of diarrhea, while supplementing 20 mg of zinc reported no reduction in episodes of diarrhea or its duration. For tuberculosis, vitamin D3 supplementation did not lower the risk of tuberculosis infection or tuberculosis disease, though it did improve height. Multiple micronutrient supplementation did not have any effect on anthropometric indices and chest radiograph improvement, however it led to an increase in height gain. Peer counseling along with contingency contract and peer-counseling alone led to a significantly higher treatment completion rate, whereas shorter course combination therapy (4 months compared with 9 months) led to increased compliance. Increasing general awareness and training of health care workers regarding tuberculosis detection also led to a 3-fold increase in tuberculosis detection. For HIV, a continuation of cotrimoxazole therapy after 96 weeks of commencing ART reduced the risk of tuberculosis and hospitalizations for malaria or infections not related to malaria. For measles, zinc supplementation did not provide any additional benefit, whereas prophylactic treatment with cotrimoxazole led to reduced measles complications and fewer pneumonia cases. For ARI, milk fortified with vitamin D3 significantly reduced the risk of ARIs. For malaria, training CHWs to give out ACT based on RDT results suggest a significant reduction in ACT prescription. For UTI, once daily 1 mg/kg of zinc supplementation in addition to standard treatment led to a faster recovery with a positive effect in reducing symptoms such as dysuria and urine frequency.

The findings from this systematic review suggest that water filtration using different filters and the provision of sanitation facilities proved to be effective in reducing diarrheal illness. These findings are in concordance with another review suggesting that clean water supply, handwashing, and sanitation reduce diarrheal deaths.⁴⁶  Another systematic review on WASH interventions among children 0 to 5 years reported that water filtration, water disinfection, and hygiene education with soap provision can all be effective in reducing diarrheal illnesses.⁴⁷  However, because of the low quality of evidence and high heterogeneity, they recommended further research in this domain.⁴⁷  Moreover, the focus of this review was children under 5 years of age, but the target population for our review was school-aged children and adolescents. Zinc supplementation for treating diarrhea has been reported to be effective among children under 5 years of age.^46,48  Our systematic review targeted school-aged children and adolescents and suggested that supplementation with 30 mg of zinc is effective in reducing the duration of diarrhea. For tuberculosis, our findings are consistent with other reviews assessing the impact of the Directly Observed Therapy program for the treatment of tuberculosis along with short message service alerts, patient education and counseling, psychological interventions, reminders and tracers, and digital health technologies and suggesting improved tuberculosis treatment outcomes and compliance.⁴⁹  Findings from another review on training sessions for nurse practitioners in the diagnosis of tuberculosis has previously reported no clear effect on improved tuberculosis detection;⁵⁰  however, the study included in our review suggested improved case detection.

There are a few limitations of our review. First, there were only a small number of studies conducted among the age group of our interest. Second, the age groups varied within each study, and third, various measures were reported for similar outcomes. Based on these limitations, we could not perform meta-analysis for many interventions and outcomes under each disease category except for WASH interventions on diarrhea. Increased focus on morbidity and mortality among children under 5 years of age has led to more evidence on interventions for the under 5 age group and consequently a dearth of information on what works among school-aged children and adolescents. Moreover, we did not find any study focusing on typhoid or paratyphoid, meningitis, syphilis, and hepatitis, which remain the leading causes of morbidity and mortality in low and middle-income countries among this age group. Future studies are needed to assess how the recommended interventions work for children and adolescents above 5 years of age, as this age group provides a second window of opportunity to target health interventions to obtain long-term sustainable impacts. The World Health Organization has guidelines for the management and research gaps for common conditions among children aged 2 to 59 months, such as pneumonia, meningitis, diarrhea, and dysentery. Many of these common conditions overlap with the high-burden infectious diseases among school-aged children and adolescents. Recommendations for pneumonia, HIV, dysentery, meningitis, and diarrhea can be adapted and modified to target children over 5 years of age.⁸  However, further evidence is needed to assess the effectiveness of certain interventions that improve detection of high-burden infectious diseases and management of these conditions at home, basic health units, or tertiary care hospitals among school-aged children and adolescents. Moreover, future studies need to focus on the relative effectiveness and duration of various interventions, including micronutrient supplementation and prophylactic antibiotic administration for conditions like HIV and measles among school-aged children and adolescents aged 5 to 19 years.

Findings from our review imply that a few interventions, including WASH, micronutrient supplementation, and prophylactic antibiotics, might be effective for high burden infectious diseases among school-aged children and adolescents; however, we need more data on the interventions to detect and manage these high-burden diseases among 5 to 19 year old children and adolescents.

ARI

acute respiratory infection

ART

anti-retroviral treatment

CHW

community health worker

HIV

human immunodeficiency virus

ILI

influenza-like illness

RCT

randomized controlled trials

UTI

urinary tract infection

WASH

water, sanitation, and hygiene