Intervention Review

You have full text access to this OnlineOpen article

Mass drug administration for malaria

  1. Eugenie Poirot1,2,
  2. Jacek Skarbinski1,
  3. David Sinclair3,
  4. S Patrick Kachur1,
  5. Laurence Slutsker1,
  6. Jimee Hwang1,2,*

Editorial Group: Cochrane Infectious Diseases Group

Published Online: 9 DEC 2013

Assessed as up-to-date: 28 FEB 2013

DOI: 10.1002/14651858.CD008846.pub2


How to Cite

Poirot E, Skarbinski J, Sinclair D, Kachur SP, Slutsker L, Hwang J. Mass drug administration for malaria. Cochrane Database of Systematic Reviews 2013, Issue 12. Art. No.: CD008846. DOI: 10.1002/14651858.CD008846.pub2.

Author Information

  1. 1

    Centers for Disease Control and Prevention, Malaria Branch, Atlanta, GA, USA

  2. 2

    University of California San Francisco, Global Health Group, San Francisco, USA

  3. 3

    Liverpool School of Tropical Medicine, Department of Clinical Sciences, Liverpool, UK

*Jimee Hwang, gdq1@cdc.gov.

Publication History

  1. Publication Status: New
  2. Published Online: 9 DEC 2013

SEARCH

 
Characteristics of included studies [ordered by study ID]
Archibald 1960 NGA

MethodsDates of study: 1957-1959

Location of study: Nigeria

Malaria endemicity (prevalence): Intervention group 1 (Arugungu - June 1958): 28% in children 1-10 years; 29% in children 0-15 years [Moderate]. Intervention group 1 (Gulmare and Koei - October 1957): 64% in children 1-10 years; 58.3% in children 0-15 years [High].

Transmission season: June to October

Malaria species: P. falciparum, P. malariae

Vector species: A. gambiae, A. funestus

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 10,000

Intervention group 2 (mean): 1300


InterventionsIntervention group 1 (Arugungu): MDA to all persons with chloroquine 600 mg and pyrimethamine 25 mg given monthly from June to October 1958. Coverage not specified. Co-intervention with IRS.

Intervention group 2 (Gulmare and Koei): MDA to all persons with chloroquine 600 mg and pyrimethamine 25 mg given every six months (November 1957, May 1958, November 1958 and March 1959). Coverage not specified. Co-intervention with IRS.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

No adverse event surveillance conducted

Adverse events reported: "There were substantial difficulties with toddlers taking chloroquine and a number of them vomited that drug."


NotesMDA added to IRS programme. The outcomes for intervention groups 1 and 2 were assessed in a sub-sample of the treated population. A third intervention group received only pyrimethamine 25 mg but was not included in the meta-analysis due to reports of rapid development of resistance.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskThe highest number of confirmed absentees reported by the investigators in September 1958 in Argungu was only 625 (6%).

Selective reporting (reporting bias)High riskThe number of children examined varied greatly between surveys without any explanation and a very small number of children were examined in Arugungu.

Other biasHigh riskAnecdotes of ill effects began to circulate and there was evidence of 'palming' of tablets.

Cavalie 1962 CMR

MethodsDates of study: 1960-1961

Location of study: Cameroon

Malaria endemicity (prevalence): Intervention group 1: 20% in children 2-9 years [Moderate]; 13% in all ages. Intervention group 2: 76% in children 2-9 years; 65% in all ages [High].

Transmission season: May to June, November to December

Malaria species: P. falciparum, P. malariae

Vector species: A. gambiae, A. funestus

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: Ages > 3 months

Sample size

Intervention group 1 (mean): 22,500

Intervention group 2 (mean): 7000


InterventionsIntervention group 1 (Secteur Sud): MDA administered to all persons aged > 3 months with chloroquine 600 mg and pyrimethamine 50 mg once for two rounds in July and November 1960. Coverage 76-92%. Co-intervention with IRS using DDT.

Intervention group 2 (Secteur Nord): MDA administered to all persons aged > 3 months with chloroquine 600 mg and pyrimethamine 50 mg once for one round in November 1960. Coverage approximately 100%. Co-intervention with IRS using DDT.


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesData presented in Table XV was used in the meta-analysis. Parasitaemia prevalence results only presented for children > 3 months to 9 years of age; meta-analysis includes only first round data. Only 13 mixed infections of P. falciparum and P. malariae were found. The remaining were P. falciparum infections only.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient reporting of attrition/exclusions to permit judgement. No reasons for missing data provided.

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasLow riskNo other bias detected

Comer 1971 PAN

MethodsDates of study: 1965-1968

Location of study: Panama

Malaria endemicity (prevalence): 17.4% in all ages [Moderate]

Transmission season: Rainy season late May to late December

Malaria species: P. falciparum, P. vivax

Vector species: Not specified

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: Ages > 6 months

Sample size

Intervention group 1 mean (range): 1709 (1548 - 1908)


InterventionsIntervention group 1 (Valle del Rio Sambu): MDA to all persons aged > 6 months with pyrimethamine 50 mg (cycles 1-25)/ 75 mg (cycles 26-49) and primaquine 40 mg given every 2 weeks for 2 years from August 1966 to April 1968. Coverage 61-87%. No co-interventions.


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

Adverse events reported: The acceptance of drugs by the population was excellent. Complaints of nausea and headache were reported, but no other serious side effects were described. None of the people who complained of headaches or nausea refused to take the medicine in subsequent cycles. The number of people who refused to take the medicine was < 1% of the population covered by the programme.


NotesNo post-intervention data


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskCoupon system used to track patients; all persons included in the surveys.

Selective reporting (reporting bias)Low riskAll intended outcomes reported

Other biasLow riskNo other bias detected

Cáceres Garcia 2008 VEN

MethodsDates of study: 2002-2007

Location of study: Venezuela

Malaria endemicity (incidence): 22/1000 monthly incidence in all ages

Transmission season: November

Malaria species: P. vivax

Vector species: Not specified

Study design: Uncontrolled before-and-after study

Evaluation design: Passive surveillance


ParticipantsAge groups included: Ages > 6 months; non-pregnant

Sample size

Intervention group 1: 25,722


InterventionsIntervention group 1 (6 municipalities in Estado Sucre): MDA to all non-pregnant persons aged >6 months with chloroquine 25 mg/kg administered over 3 days and primaquine 3.5 mg/kg administered over 7 days in November 2002. Coverage 77% (of census)/ 86% (of included). No co-intervention specified.


OutcomesParasitaemia incidence

No adverse event surveillance conducted

No adverse events reported


NotesMDA done in setting of an outbreak


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskPassive surveillance of large municipalities after one round of treatment

Selective reporting (reporting bias)Low riskNo evidence of selective reporting

Other biasLow riskNo other bias detected

De Zulueta 1961 UGA

MethodsDates of study: 1959-1960

Location of study: Uganda

Malaria endemicity (prevalence): 34% in children 2-9 years; 17% in all ages [Moderate]

Transmission season: Rainy season April to May, August to November

Malaria species: P. falciparum, P. malariae

Vector species: A. gambiae, A. funestus

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 mean (range): 30,384 (10,303 - 59,605)


InterventionsIntervention group 1 (North Kigezi): MDA administered to all persons with chloroquine 600 mg and pyrimethamine 50 mg every three months for four rounds at the time of IRS application from May 1959 to May 1960. Coverage 80%. Co-intervention with IRS.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesOutcomes assessed in a sub-sample of the treated population.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskCooperation of the local inhabitants was remarkably good and not a single dwelling was left unsprayed

Selective reporting (reporting bias)Low riskIncreased number of samples from hyperendemic areas in the post-intervention survey

Other biasLow riskNo other bias detected

De Zulueta 1964 UGA

MethodsDates of study: 1960

Location of study: Uganda

Malaria endemicity (prevalence): 23% in children 2-9 years; 21% in all ages [Moderate]

Transmission season: Rainy season April to May, August to November

Malaria species: P. falciparum, P. malariae

Vector species: A. gambiae

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: Ages > 3 months

Sample size

Intervention group 1 (mean): 16,000


InterventionsIntervention group 1 (Lake Bunyonyi): MDA to all persons aged > 3 months with chloroquine 600 mg and pyrimethamine 50 mg once per round for two rounds (April to May 1960 and September to October 1960). Coverage approximately 50% in the first round. Co-intervention with IRS.


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesOutcomes assessed in a sub-sample of the treated population. A. funestus disappeared after one year of spraying and no new malaria cases were noted two years later.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient detail, but the total number surveyed differs greatly between surveys

Selective reporting (reporting bias)Low riskAll relevant outcomes were measured

Other biasHigh riskOnly about half of the population was given MDA during the first round

Escudie 1962 BFA

MethodsDates of study: 1960-1961

Location of study: Burkina Faso

Malaria endemicity (prevalence): Comparison group 1: 56.1% in children 0-10 years [High]

Transmission season: June to December

Malaria species: P. falciparum, P. ovale, P. malariae

Vector species: A. gambiae, A. funestus, A. nili

Study design: Non-randomized controlled study

Evaluation design: Cross-sectional surveys


ParticipantsAge group included: All ages

Sample size

Intervention group 1 (mean): 1890

Intervention group 2 (mean): 2560

Intervention group 3 (mean): 5400

Intervention group 4 (mean): 3490

Comparison group 1 (mean): Not described

Comparison group 2 (mean): Not described


InterventionsIntervention group 1: MDA to all persons with a single dose of either chloroquine-primaquine (600 mg/15 mg) or amodiaquine-primaquine (600 mg/15 mg) every 28 days from June to December 1960. Coverage 75.2 to 91.2%. No co-interventions.

Intervention group 2: MDA to all persons with a single dose of either chloroquine-primaquine (600 mg/15 mg) or amodiaquine-primaquine (600 mg/15 mg) every 14 days from June to December 1960. Coverage 84.1 to 96.5%. No co-interventions.

Intervention group 3: MDA to all persons with a single dose of either chloroquine-primaquine (600 mg/15 mg) or amodiaquine-primaquine (600 mg/15 mg) every 28 days from June to December 1960. Coverage 80.9 to 91.8%. Co-intervention with IRS using DDT annually.

Intervention group 4: MDA to all persons with a single dose of either chloroquine-primaquine (600 mg/15 mg) or amodiaquine-primaquine (600 mg/15 mg) every 14 days from June to December 1960. Coverage 82.1 to 93.8%. Co-intervention with IRS using DDT annually.

Comparison group 1: Control villages. No co-interventions.

Comparison group 2: Villages sprayed with IRS using DDT annually. No other co-interventions.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesOutcomes assessed in a sub-sample of the treated population (children 0-10 years). Baseline data from June 1960 survey. Ninety percent of cases are P. falciparum infections; P. ovale is rare and P. malariae is very rare.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskAssignment to MDA was not randomized although drug assignment was randomized

Allocation concealment (selection bias)High riskNon-randomized controlled study

Baseline imbalance (selection bias)High riskBaseline parasitaemia estimates are not balanced between the intervention groups and the comparison groups. Also, there was large variability in endemicity between comparison group 1 villages.

Contamination protectionUnclear riskInsufficient information to permit judgement

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskParticipants and personnel aware of treatment, but unclear if this impacted outcomes

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot blinded, but unclear if this impacted outcomes

Incomplete outcome data (attrition bias)
All outcomes
Low riskAdults included in MDA, but not in the evaluation. Only children 0-10 years of age were examined in the malaria surveys before, during and after MDA.

Selective reporting (reporting bias)Low riskAll pre-specified outcomes of interest are reported

Other biasUnclear riskAtypical seasonal changes experienced in 1959-1960, but it is unclear if these changes impacted outcomes.

Gabaldon 1959 VEN

MethodsDates of study: 1956-1957

Location of study: Venezuela

Malaria endemicity (incidence): 0.4/1000 baseline monthly incidence

Transmission season: May to November

Malaria species: P. vivax

Vector species: A. aquasalis, A. nuneztovari

Study design: Uncontrolled before-and-after study

Evaluation design: Active and passive surveillance


ParticipantsAge groups included: Ages > 1 month

Sample size

Intervention group 1 (mean): 111,995


InterventionsIntervention group 1: Eastern Venezuela (174 localities, 3084 houses, 16,416 persons) and Western Venezuela (735 localities, 17,638 houses, 95,579 persons): MDA to all persons aged > 1 month with pyrimethamine 50 mg per week for 24 weeks from July 1957 to December 1957. Coverage not specified. Co-intervention with IRS.


OutcomesParasitaemia incidence

No adverse event surveillance conducted

No adverse events reported


NotesMDA added to IRS program


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll houses numbered. Envelope system for drug dispensers and slide collectors. Cooperation of the people was excellent. Active search for all infections and passive search at all medical dispensaries in the area.

Selective reporting (reporting bias)Low riskMost persons received more than 19 treatments; however, the actual figures are not reported due to "lack of mechanical tabulation of the data". The number of persons with relapses who had less than 19 treatments demonstrated similar trends to those who received 19 or more treatments.

Other biasLow riskNo other bias detected

Garfield 1983 NIC

MethodsDates of study: 1981-1982

Location of study: Nicaragua

Malaria endemicity (incidence): 0.4/1000 baseline monthly incidence

Transmission season: November to March

Malaria species: P. falciparum, P. vivax

Vector species: A. albimanus

Study design: Uncontrolled before-and-after study

Evaluation design: Passive surveillance


ParticipantsAge groups included: Ages > 1 year

Sample size

Intervention group 1 (mean): 2,300,000


InterventionsIntervention group 1: MDA administered to all persons aged > 1 year with chloroquine 1500 mg and primaquine 45 mg over three days given once to the entire population of Nicaragua in November 1981. Coverage 70-80%. Co-intervention with larviciding using large scale application of temephos to peridomiciliary breeding sites targeting Aedes aegypti, but likely to have an effect on anophelines.


OutcomesParasitaemia incidence

No adverse event surveillance conducted

Adverse events reported: Common side effects included dizziness, nausea, vomiting and diarrhoea. Occasional cases of psychomotor disturbance, temporary psychological abnormalities and haemolysis.


NotesData used in the meta-analysis was extrapolated from graphs presented in the text; baseline MDA estimates were determined using monthly surveillance data from 1974-1981.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskSingle treatment episode after conducting a census, door-to-door education and promotion of community participation.

Selective reporting (reporting bias)Low riskNational passive surveillance

Other biasLow riskNo other bias detected

Gaud 1953 MAR

MethodsDates of study: 1952

Location of study: Morocco

Malaria endemicity (prevalence): 41.5% in all ages (baseline) [High]

Transmission season: June to October

Malaria species: P. falciparum, P. vivax

Vector species: Not specified

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size:

Intervention group 1 (mean): 3000


InterventionsIntervention group 1: MDA administered to all persons with amodiaquine 600 mg given once in the summer of 1952. Coverage not specified. No co-interventions.


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

No adverse events reported


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information to permit judgement

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasLow riskNo other bias detected

Hii 1987 MYS

MethodsDates of study: 1984-1985

Location of study: Malaysia

Malaria endemicity (prevalence): Intervention group 1 (December 1984 baseline survey): 46.3% in children 0-8 years [High]; Intervention group 2 (December 1984 baseline survey): 55.6% in children 0-8 years [High]

Transmission season: Perennial

Malaria species: P. falciparum, P. malariae, P. vivax

Vector species: A. balabacensis

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys and active surveillance


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 754

Intervention group 2 (mean): 148


InterventionsIntervention group 1: MDA administered to all persons (139 households in five villages) with sulfadoxine-pyrimethamine (1500 mg/75 mg) and primaquine 30 mg once in December 1984 to January 1985. Coverage 87%. Co-intervention with permethrin-impregnated bed nets to all households.

Intervention group 2: MDA administered to all persons (nine households in one village) with sulfadoxine-pyrimethamine (1500 mg/75 mg) and primaquine 30 mg once in December 1984 to January 1985. Coverage 76%. No co-interventions.


OutcomesParasitaemia prevalence

Parasitaemia incidence

Gametocytaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesThough the entire population was treated, thick and thin blood films were collected during eight surveys on a population of 286 children aged 0-8 years. Only data for these children were reported and therefore used in the meta-analysis. Furthermore, because the study design included a comparison group that received MDA, the intervention and comparison groups will be treated as two intervention groups and each intervention group will be analyzed in the meta-analysis as a separate uncontrolled before-and-after study. Lastly, due to insufficient information to extract incidence data, parasitaemia incidence was not included as an outcome in the meta-analysis.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
High riskThough the entire population was treated, thick and thin blood films were collected during eight surveys on a population of 286 children aged 0-8 years. Only 29.7% of children were present at every one of the eight sessions.

Selective reporting (reporting bias)High riskThe study report fails to include results on P. vivax infections that would be expected to have been reported for such a study. The study methods indicate that thick blood films will be classified as "positive or negative for asexual and/or sexual parasites of either P. falciparum, P. vivax, P. malariae, or mixed infections". Only parasitological findings for P. falciparum are described and presented in detail.

Other biasLow riskWith the exception of two study villages, which are both intervention group 1 sites, the study villages are "well separated and demarcated". Therefore, it is unlikely that contamination between sites occurred. All villages also received the same treatment dose and schedule. However, it should be noted that in the meta-analysis, the two interventions were analyzed as two separate uncontrolled before-and-after studies.

Houel 1954 MAR

MethodsDates of study: 1953

Location of study: Morocco

Malaria endemicity (prevalence): 14.3%, children only (August 1953 baseline survey) [Moderate]

Transmission season: July to November

Malaria species: P. falciparum, P. malariae, P. vivax

Vector species: Not specified

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 9999


InterventionsIntervention group 1: MDA administered to all persons with pyrimethamine 100 mg once in June 1953 to September 1953. Coverage not specified. Co-intervention with IRS prior to MDA.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesOnly results from the 147 children examined were included in the meta-analysis.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
High riskWhile adults were included in MDA, only a subset of children were included in the evaluation.

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasHigh riskNo data on coverage of intervention

Jones 1954 KEN

MethodsDates of study: 1952-1953

Location of study: Kenya

Malaria endemicity (prevalence): 34.8% (baseline survey in a random sample of adults and infants); 32.6% (baseline survey in school children) [Moderate]

Transmission season: January to March, May to August

Malaria species: P. falciparum, P. malariae

Vector species: A. gambiae, A. funestus

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 3721 (including 297 school children)


InterventionsIntervention group 1: MDA administered to all persons in Makueni with pyrimethamine 100 mg once for three rounds in September 1952, March 1953 and September 1953. Coverage not specified. No co-interventions.


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesFollowing the first MDA round, blood smears were taken from random samples of the adult and infant (< 5 years) population and from all school children for a year. Due to the high degree of resistance that developed following two MDA rounds, parasitaemia prevalence results in the meta-analysis reflect only first round MDA results for infants and adults.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskIndividual data kept of all school children and of all subjects with malaria attending the dispensary

Selective reporting (reporting bias)Low riskBlood smears collected from random samples of adults and infants and of all school children monthly for a year following the first MDA round. All pre-specified outcomes have been reported.

Other biasHigh riskComplicated by resistance

Jones 1958 KEN

MethodsDates of study: 1952-1953

Location of study: Kenya

Malaria endemicity (prevalence): Intervention group 1 (September 1952): 60% in school-age children; Comparison group 1 (September 1953): 34% in school-age children [Moderate]

Transmission season: January to March, May to August

Malaria species: P. falciparum, P. malariae, P. vivax

Vector species: A. gambiae, A. funestus

Study design: Non-randomized controlled study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages; school-age children

Sample size

Intervention group 1 (range): 3721-4500

Comparison group 1: Not specified


InterventionsIntervention group 1: MDA administered to all school children in Makueni with pyrimethamine 100 mg for three rounds in September 1952, March 1953 and September 1953. Coverage not specified. No co-interventions.

Comparison group 1: School children in Okia used as a comparison arm. No co-interventions.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesOutcome data for the intervention group is a subset of the Jones 1954 KEN study. The meta-analysis only included first-round results. Gametocytaemia prevalence data is forP. falciparum only.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNon-randomized controlled study

Allocation concealment (selection bias)High riskNon-randomized controlled study

Baseline imbalance (selection bias)High riskBaseline parasitaemia estimates are not balanced between the intervention group and the comparison group.

Contamination protectionUnclear riskAlthough the comparison group site was 13 miles from the intervention group site, there is no indication whether the control group was adequately protected against contamination. It is quite possible that the control group received the intervention.

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskParticipants and personnel aware of treatment, but unclear if this impacted outcomes

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot blinded, but unclear if this impacted outcomes

Incomplete outcome data (attrition bias)
All outcomes
Low riskIndividual data kept of all school-age children and of all subjects with malaria attending the dispensary. No antimalarials were sold in local shops. At the end of the 12th month of evaluation, 221 children remained out of the original 297 children.

Selective reporting (reporting bias)Low riskBlood smears from random samples and all school-age children. Over the course of the study, the school population rose by 178 children. To avoid confusion, the investigators excluded these additional children from the figures used to compile prevalence and only reported data from the original 297 children.

Other biasHigh riskComplicated by drug resistance

Kaneko 2000 VUT

MethodsDates of study: 1991-1999

Location of study: Vanuatu

Malaria endemicity (prevalence): Intervention group 1 (January - September 1991): 15.7% in all ages; Comparison group 1 (May 1990): 28.8% in all ages [Moderate].

Transmission season: December to April

Malaria species:P. falciparum, P. vivax

Vector species: A. farauti

Study design: Non-randomized controlled study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 718

Comparison group 1 (mean): 19,289


InterventionsIntervention group 1: MDA administered to all persons in Aneityum weekly for nine weeks with chloroquine 600 mg and sulfadoxine-pyrimethamine 1500 mg/75 mg and primaquine 45 mg once a week in weeks 1, 5, and 9; chloroquine 300 mg and primaquine 45 mg once a week in weeks 2, 3, 4, 6, 7, and 8 in September 1991 to November 1991. Coverage 79 to 92%. Co-intervention with larvivorous fish in several identified breeding sites and universal coverage with insecticide treated bed nets (about 0.94 nets per villager).

Comparison group 1: Persons living in Malakula Island. Co-intervention with bed nets (approximately 20% coverage).


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

Adverse events reported: Some villagers reported vomiting after taking the tablets.


NotesAnother village on Futana island was included in the study for comparison; however, because no parasitaemia was detected in the two surveys on Futuna, it was excluded from the meta-analysis. The meta-analysis only included data from Aneityum for the months of January and September 1991 (before MDA) and March 1998 (post-MDA).


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNon-randomized controlled study

Allocation concealment (selection bias)High riskNon-randomized controlled study

Baseline imbalance (selection bias)Low riskAccording to investigators, "the parasite rates were initially similar on Aneityum and Malakula islands and in general, decreased with age".

Contamination protectionLow riskThe comparison group was a village from Malakula, an adjacent island; therefore, it is unlikely that the comparison group received the intervention.

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNot blinded, but unclear if this impacted outcomes

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot blinded, but unclear if this impacted outcomes

Incomplete outcome data (attrition bias)
All outcomes
Low riskOnly 7.9% of doses unable to be administered and only 3.8% doses were not properly reported and could not be confirmed. The overall calculated compliance rate of the remaining doses was 88.3%.

Selective reporting (reporting bias)High riskOf the 13 surveys, two covered only school children whereas the other 11 surveys covered the entire population of Aneityum.

Other biasLow riskNo other bias detected

Kligler 1931 PSE

MethodsDates of study: 1930

Location of study: Palestine (known as British Mandate Palestine at the time of the study's publication)

Malaria endemicity (prevalence): 35% in all ages; 67% in children 2-10 years [High]

Transmission season: October to December

Malaria species:P. falciparum, P. malariae, P. vivax

Vector species: A. elutus

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys and active surveillance


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 mean (range): 953 (899-993)


InterventionsIntervention group 1: MDA administered to all persons in five selected villages with plasmochine 30 mg plus quinine 900 mg twice daily for five days every three weeks for three rounds between September and November 1930. Coverage 78.8%. No co-interventions.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

Adverse event surveillance conducted (active during the course of the treatment)

Adverse events reported: No ill results were noted during the entire course of treatment.


NotesNoted that repeated treatments tended to increase resistance.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
High riskThere was a large drop in the number of villages surveyed from baseline to post-survey without any explanation.

Selective reporting (reporting bias)High riskFive villages were treated but only select villages reported outcome data.

Other biasLow riskNo other bias detected

Kondrashin 1985 IND

MethodsDates of study: 1981

Location of study: India

Malaria endemicity (incidence): 4/1000 baseline monthly incidence

Transmission season; April to August

Malaria species: P. falciparum, P. vivax

Vector species: Not specified

Study design: Uncontrolled before-and-after study

Evaluation design: Passive surveillance


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 51,325


InterventionsIntervention group 1: MDA administered to all persons with chloroquine 600 mg (plus primaquine 45 mg in falciparum areas only) for one round in March to May 1981 in four primary health centres and two rounds in February to March 1981 and June to September 1981 in four other primary health centres. Coverage 85%. Co-intervention with IRS.


OutcomesParasitaemia incidence

No adverse event surveillance conducted

No adverse events reported


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskOnly 1 or 2 rounds of treatment with 85% coverage

Selective reporting (reporting bias)Unclear riskNo mention of the thoroughness of passive surveillance

Other biasHigh riskA likely increase inP. falciparum due to labour movement into treated area

Malaria_Taiwan 1991 TWN

MethodsDates of study: 1955

Location of study: Taiwan

Malaria endemicity (prevalence): 4.12% in all ages (May 1955 survey); 2.93% in all ages (November 1955) [Low]

Transmission season: Not described

Malaria species: P. falciparum, P. malariae, P. vivax

Vector species: A. maculatus, A. minimus, A. sinensis

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys and passive surveillance


ParticipantsAge groups included: All ages, except infants

Sample size

Intervention group 1 mean (range): 1520 (1502-1537)


InterventionsIntervention group 1: MDA administered to all persons, except infants, in Lanyu with a single dose of chloroquine (12 mg/kg) in November 1955. Coverage not specified. Co-intervention with IRS using DDT.


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesPost-MDA (> 12 months) estimated using survey data from April-May 1957 and April 1960


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient reporting of attrition/exclusions to permit judgement

Selective reporting (reporting bias)Low riskThe first three malariometric baseline surveys reported consisted of only a portion of the entire population on the island. Subsequent surveys examined all inhabitants. While these disproportionate samples could result in a certain bias when compared to the remaining surveys that comprised the entire population, the investigators weighted the first three surveys according to the natural distribution of the population.

Other biasLow riskNo other bias detected

Metselaar 1961 PNG

MethodsDates of study: 1958-1959

Location of study: Papua New Guinea

Malaria endemicity (prevalence): 46-80% in children 2-11 years; 46% in all ages before spraying [High]; During spraying 13-21% in children 2-11 years; 12% in all ages [Moderate]

Transmission season: Not described

Malaria species: P. falciparum, P. malariae, P. vivax

Vector species: A. punctulatus, A. farauti, A. koliensis

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 2500


InterventionsIntervention group 1 (Sentani): MDA administered to all persons in sprayed areas with chloroquine 450 mg plus pyrimethamine 50 mg at weekly intervals for five rounds in 1958 and for one round in 1959. Two villages with high absolute parasite rates received an additional round of treatment in 1959. In addition, during all rounds, positives received chloroquine for an additional three successive days, completing a full course (1350 mg base for adults). Coverage 90%. Co-intervention with IRS.


OutcomesParasite prevalence

No adverse event surveillance conducted

No adverse events reported


NotesBaseline data from 1958 survey


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low risk90% coverage, but no further description

Selective reporting (reporting bias)Unclear riskSelection for inclusion in surveys not described

Other biasLow riskNo other bias detected

Molineaux 1980 NGA

MethodsDates of study: 1970-1975

Location of study: Nigeria

Malaria endemicity (prevalence): 46% in all ages [High]

Transmission season: April to October

Malaria species: P. falciparum, P. malariae, P. ovale

Vector species: A. gambiae, A. funestus

Study design: Non-randomized controlled study

Evaluation design: Cross-sectional surveys and active surveillance


ParticipantsAge groups included: All ages, but infants not included in MDA until their first malaria episode.

Sample size

Intervention group 1 (mean): 14,129

Intervention group 2 (mean): 1810

Comparison group 1 (mean): 32,828

Comparison group 2 (mean): ND


InterventionsIntervention group 1 (Low frequency MDA+IRS group): MDA administered to all ages, except for infants who have not had their first malaria episode, with sulfalene-pyrimethamine 500 mg/25 mg every 10 weeks from April 1972 to October 1973. Coverage 85%. Co-intervention with IRS using propoxur 3-4 rounds per year.

Intervention group 2 (High frequency MDA+IRS group): MDA administered to all ages, except for infants who have not had their first malaria episode, with sulfalene-pyrimethamine 500 mg/25 mg every two weeks during the wet season and every 10 weeks during the dry season from April 1972 to October 1973. Coverage 85%. Co-intervention with IRS using propoxur 3-4 rounds per year.

Comparison group 1: IRS using propoxur 3-4 rounds per year.

Comparison group 2: No interventions.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

Mortality

No adverse event surveillance conducted

No adverse events reported


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNon-randomized controlled study

Allocation concealment (selection bias)High riskNon-randomized controlled study

Baseline imbalance (selection bias)Low riskSimiliar malaria characteristics between groups

Contamination protectionLow riskIt was desirable to allocate contiguous areas to the same treatment and also to reduce the effect of migrations by having similarly treated buffer zones around the evaluation villages.

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNot mentioned but unclear if this impacted outcomes

Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent reexamination of slides

Incomplete outcome data (attrition bias)
All outcomes
Low riskOperation aimed for total coverage

Selective reporting (reporting bias)Low riskThe surveys covered the total de facto population of selected village clusters and all possible outcomes measured and reported.

Other biasLow riskNo other bias detected

Najera 1973 NGA

MethodsDates of study: 1966-1968

Location of study: Nigeria

Malaria endemicity (prevalence): Comparison group 1: 28.9% in all ages [Moderate]

Transmission season: May to September

Malaria species: P. falciparum

Vector species: A. gambiae, A. funestus

Study design: Non-randomized controlled study (no post-intervention measurements)

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: Ages > 3 months

Sample size

Intervention 1 mean (range): 52,000 (52,060 to 53,897)

Comparison 1 mean: 11,500


InterventionsIntervention group 1: MDA administered to all persons aged > 3 months with chloroquine 450 mg and pyrimethamine 45 mg every 60 days for 11 rounds from November 1966 to August 1968. Coverage 78 to 92%. Co-intervention with IRS.

Comparison group 1: Co-intervention with IRS only. Coverage not described.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

Active adverse event surveillance conducted

Adverse events reported: Direct observation of 5003 treatments during MDA rounds 9 and 10 revealed 2% vomiting immediately after taking the drug. When a subset of the population was asked about vomiting, 9% reported this symptom.


NotesData collected during rounds 2 to 11 are summarized as during MDA results. This is problematic as the initial decline and later rise of cases during the two years of drug administration is aggregated. Evaluation conducted in a subset of treated population.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNon-randomized controlled study

Allocation concealment (selection bias)High riskNon-randomized controlled study

Baseline imbalance (selection bias)High riskThe comparison area was not comparable to the intervention area in terms of entomologic or parasitological parameters.

Contamination protectionLow riskTreated large peripheral zone, but evaluation done in central zone only

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNot blinded, but unclear if this impacted outcomes

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot mentioned, but unclear if this impacted outcomes

Incomplete outcome data (attrition bias)
All outcomes
Low riskRecorded census and population movement without large loss to follow-up

Selective reporting (reporting bias)Low riskRandom sampling of clusters of 200 people for the parasitological surveys

Other biasLow riskNo other bias detected

Paik 1974a SLB

MethodsDates of study: 1972

Location of study: Solomon Islands (known as British Solomon Islands at the time of the study's publication)

Malaria endemicity (prevalence): 27.8% all ages (May 1972 survey) [Moderate]

Transmission season: Rainy season December to April

Malaria species: P. falciparum, P. vivax

Vector species:A. farauti

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys, passive surveillance and active surveillance


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): Not specified


InterventionsIntervention group 1 (Nggela archipelago): MDA administered to all persons with chloroquine 600 mg and pyrimethamine 50 mg monthly for four months from July to October 1972. Coverage 90%. Co-intervention with IRS.


OutcomesParasitaemia prevalence (includes both passive and active case detection for the period during and after the intervention)

Parasitaemia incidence (population size not given)

No adverse event surveillance conducted

No adverse events reported


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient reporting of attrition/exclusions to permit judgement

Selective reporting (reporting bias)High riskOnly 50% of children 2-9 years old included in the pre-MDA and post-MDA household surveys

Other biasHigh riskBaseline surveillance did not include active case detection

Paik 1974b SLB

MethodsDates of study: 1972-1973

Location of study: Solomon Islands (known as British Solomon Islands at the time of the study's publication)

Malaria endemicity (incidence): 15/1000 baseline monthly incidence

Transmission season: Rainy season December to April

Malaria species: P. vivax, P. malariae

Vector species: A. farauti

Study design: Uncontrolled before-and-after study

Evaluation design: Passive surveillance


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 1200


InterventionsIntervention group 1 (Wagina and Shortland): MDA administered to all persons with chloroquine 1500 mg and primaquine 75 mg over five days every three months for three rounds from October 1972 to March 1973. Coverage 90%. No co-interventions.


OutcomesParasitaemia incidence

No adverse event surveillance conducted

No adverse events reported


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information to permit judgement

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasLow riskNo other bias detected

Ricosse 1959 BFA

MethodsDates of study: 1958-1959

Location of study: Burkina Faso

Malaria endemicity (prevalence): Intervention group 1 (March-May 1958 baseline survey): 15.3% in children 0-9 years [Moderate]; Intervention group 2 (March to May 1958 baseline survey): 56.0% in children 0-9 years [High]

Transmission season: June to October

Malaria species: P. falciparum, P. malariae

Vector species: A. gambiae, A. funestus

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 5000

Intervention group 2 (mean): 3000


InterventionsIntervention group 1 (Zone A): MDA administered to all persons with pyrimethamine 50 mg every two weeks for eight rounds in June to September 1958. Coverage 82-91%. Co-intervention with IRS using DDT.

Intervention group 2 (Zone B): MDA administered to all persons with pyrimethamine 50 mg every two weeks for eight rounds in June to September 1958. Coverage 82-91%. No co-interventions.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesOutcomes assessed in sub-sample of treated population (0-9 years). Data presented in Table 1 was used in the meta-analysis.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
High riskIn Zone B, pyrimethamine distribution stopped on September 20th and resumed in October, so the study was unable to follow the entire evolution of resistance that apparently began during the fourth month of distribution. Also, the method of selection of children 2-9 years is unclear. They took monthly blood samples in all children 0-23 months, but due to the large sample size selected only a proportion of children 2-9 years to examine.

Selective reporting (reporting bias)Unclear riskInsufficient information to permit judgement

Other biasHigh riskComplicated by resistance in the fourth month of MDA

Roberts 1964 KEN

MethodsDates of study: 1953-1954

Location of study: Kenya

Malaria endemicity (prevalence): 28% in 1953 [Moderate] and 22% in 1954 [Moderate] in all ages in Tiriki

Transmission season: May to July

Malaria species:P. falciparum, P. malariae

Vector species: A. gambiae, A. funestus

Study design: Non-randomized controlled study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 101,000

Intervention group 2 (mean): 99,000

Comparison group 1 (mean): Not specified

Comparison group 2 (mean): Not specified


InterventionsIntervention group 1 (Nandi District 1953): MDA administered to all persons with pyrimethamine 50 mg once in May 1953. Coverage 95%. No co-intervention.

Intervention group 2 (Nandi District 1954): MDA administered to all persons with pyrimethamine 50 mg once in May 1954. Coverage 95%. No co-intervention.

Comparison group 1 (Tiriki control area 1953): No interventions

Comparison group 2 (Tiriki control area 1954): No interventions


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesIntended to control epidemics.

In the methods, it states: "one hundred thick blood films were taken in treated and untreated areas from persons in each of the age groups 0-10 years, 11-20 years, and 21 years and older". Therefore, we assumed that the number of total patients examined was 300 for both intervention and comparison groups to determine the number of cases identified in our calculations for parasitaemia prevalence.

Outcomes were assessed in a sub-sample of the treated population.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNon-randomized controlled study

Allocation concealment (selection bias)High riskNon-randomized controlled study

Baseline imbalance (selection bias)High riskHigher baseline parasitaemia in the control area

Contamination protectionLow riskNot described but trial area was very large

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNot blinded but unclear if this impacted outcomes

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot mentioned but unclear if this impacted outcomes

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll inhabitants living in the selected area received treatment

Selective reporting (reporting bias)Unclear riskUnclear who or how many were included in the malaria surveys

Other biasLow riskNo other bias detected

Schneider 1961 BFA

MethodsDates of study: 1960-1961

Location of study: Burkina Faso

Malaria endemicity (prevalence): Comparison group 1 (baseline survey): 59.4% in children 2-9 years [High]

Transmission season: August to September

Malaria species: P. falciparum, P. vivax

Vector species: Not described

Study design: Non-randomized controlled study (no pre-intervention measurements)

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 2500

Intervention group 2 (mean): 3535

Comparison group 1 (mean): Not specified


InterventionsIntervention group 1: MDA administered to all persons with a combination of 600 mg base chloroquine or amodiaquine and 15 mg base primaquine every 14 days in June to December 1960 for 15 rounds. No co-intervention. Coverage 90%.

Intervention group 2: MDA administered to all persons with 600 mg base amodiaquine and 15 mg base primaquine every 14 days in June to December 1960 for eight rounds. Coverage not specified. Co-intervention with IRS using DDT once a year in May 1960.

Comparison group 1: Control zone free of any intervention (house spraying or treatment). Coverage not specified.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesData on children 0-9 years were reported; however, data could only be abstracted for 2-9 years to draw appropriate comparisons. In addition, data for during MDA for the intervention groups were estimated using only October 1960 survey data; during MDA data for the comparison group was only provided for October 1960.

Intervention sample size is based on the 2500 inhabitants of the three villages surveyed; half were randomized to receive amodiaquine and primaquine while the other half received chloroquine and primaquine.

A third intervention group was treated with a combination of 600 mg base chloroquine or amodiaquine and 15 mg base primaquine every 14 days in June to December 1960; however, due to lack of detailed data presented, this group was not included in the meta-analysis.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNon-randomized controlled study

Allocation concealment (selection bias)High riskNon-randomized controlled study

Baseline imbalance (selection bias)Low riskPatient outcomes were measured prior to the intervention. According to investigators, no important differences were present across study groups.

Contamination protectionUnclear riskInsufficient information to permit judgement

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskInsufficient information to permit judgement

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInsufficient information to permit judgement

Incomplete outcome data (attrition bias)
All outcomes
High riskAdults were treated during MDA, but were not included in the evaluation.

Selective reporting (reporting bias)High riskA monthly distribution schedule was also administered in the study; however due to the poor quality data, minimal results were described.

Other biasUnclear riskInsufficient information to permit judgement

Shekalaghe 2011 TZA

MethodsDates of study: 2008

Location of study: Tanzania

Malaria endemicity (prevalence): 0% in all ages [Low]

Transmission season: March to May, October to November

Malaria species: P. falciparum

Vector species: Not described

Study design: Cluster-randomized trial

Unit of randomization: Geographical clusters of households

Adjusted analyses for clustering: Yes

Adjustment method: Generalized estimating equations

ICC: Not described

Numbers of clusters: 16

Number of people: 3457

Average cluster size: 216

Evaluation design: Cross-sectional surveys, passive surveillance and active surveillance in children


ParticipantsAge groups included: Ages > 1 year, but individuals who had received a full dose of ACT in the two weeks before the intervention were excluded.

Sample size

Intervention group 1 (mean): 1110

Comparison group 1 (mean): 2347


InterventionsIntervention group 1: MDA administered to all persons in eight clusters in four villages in Lower Moshi with sulphadoxine-pyrimethamine (25 mg + 1.25 mg/kg as a single dose on the first day) plus artesunate (4 mg/kg/day for three days) plus primaquine (0.75 mg/kg as a single dose on the third day). Pregnant women received sulphadoxine-pyrimethamine (25 mg + 1.25 mg/kg 25 mg + 1.25 mg/kg as a single dose on the first day) plus amodiaquine (10 mg/kg once daily for three days). Anaemic individuals received sulphadoxine-pyrimethamine (25 mg + 1.25 mg/kg 25 mg + 1.25 mg/kg as a single dose on the first day) plus artesunate (4 mg/kg/day for three days). Coverage 93%. Co-intervention with background bed net use (25.1% to 36.1%) and a single treatment campaign for trachoma with azithromycin was undertaken by a non-governmental organisation.

Comparison group 1: Placebo administered to all persons in eight clusters once daily over three days. Coverage not described. Co-intervention with background bed net use (25.1% to 36.1%) and a single treatment campaign for trachoma with azithromycin was undertaken by a non-governmental organisation.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

Active adverse event surveillance with haemoglobin monitoring conducted in a subset of the population

Adverse events reported: One individual was diagnosed with a severe skin reaction in the week following MDA. Upon review, it was determined that the event was drug related. A second individual presented with skin hyperpigmentation on the face, which was determined unrelated to drug treatment. Both individuals were treated with steroids and monitored until symptoms disappeared. In those given primaquine, moderate anaemia (Hb level of <8 g/dL) was observed in 40% (6/15 individuals) of the G6PD A-, 11.1% (3/27 individuals) of the G6PD A, and 4.5% (18/399 individuals) of the G6PD B individuals; one case of severe anaemia (Hb level of <5 g/dL) was observed.


NotesThe prevalence outcomes were assessed in a sub-sample of the treated population.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskRandomized using computer generated randomization tables

Allocation concealment (selection bias)Low riskNot described, but low risk with the randomization of a small number of clusters presumably by the investigator

Baseline imbalance (selection bias)Low riskBaseline demographic and malaria characteristics were similar

Contamination protectionLow riskHouseholds that were located between clusters (ie within 1 km distance from the boundary of intervention and/or control clusters) were considered as buffer zones. Members of these households received the intervention in order to minimize contamination.

Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo was used in the comparison arm

Blinding of outcome assessment (detection bias)
All outcomes
Low riskThe measurement of outcomes for intervention and comparison arms were identical.

Incomplete outcome data (attrition bias)
All outcomes
Low riskHigh coverage of intervention and population movement monitored

Selective reporting (reporting bias)Low riskFor each cross-sectional survey, individuals were randomly selected from computer-generated random tables. All planned outcome measures were reported.

Other biasLow riskNo other bias detected

Simeons 1938 IND

MethodsDates of study: 1935

Location of study: India

Malaria endemicity (incidence): 156 cases/1000 baseline monthly incidence in all age groups

Transmission season: March to August

Malaria species: P. vivax

Vector species: A. culicifacies

Study design: Uncontrolled before-and-after study

Evaluation design: Passive surveillance


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 5650


InterventionsIntervention group 1 (Mill Area): MDA administered to all persons with atebrin intramuscular 300 mg daily for 2 days and plasmochin simplex 60 mg daily for three days once in May to June 1935. Coverage 100%. Co-intervention with oiling for larval control after MDA.


OutcomesParasitaemia incidence

Passive event surveillance conducted

Adverse events reported: Haemoglobinuria occurred in 4 cases (2 severe and died; 2 mild); three of the cases were from the same household and all were taking treatment for syphilis. Fatal cases known to have syphilis and unlikely to be associated with atebrin; although potentially associated with plasmochin. Abcesses reported in 49 small children and weak adults. "Giddiness" reported with atebrin.


NotesBaseline monthly incidence was estimated using survey data from May 1934 to April 1935 prior to MDA. Data used in the meta-analysis was extrapolated from graphs presented in the text.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
Low riskEvery person in the Mill Area was treated; extensive propaganda was carried out to bring every fever case to the doctor.

Selective reporting (reporting bias)Low riskPassive surveillance data for the entire population was reported

Other biasLow riskNo other bias detected

Singh 1953 IND

MethodsDates of study: 1952-1953

Location of study: India

Malaria endemicity (prevalence): 22% in all ages [Moderate]

Transmission season: September to November

Malaria species: P. falciparum

Vector species: Not described

Study design: Non-randomized controlled study

Evaluation design: Cross-sectional surveys and active surveillance


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 125

Comparison group 1 (mean): 55

Comparison group 2 (mean): 121


InterventionsIntervention group 1: MDA administered to all persons with amodiaquine 600 mg every two weeks for ten weeks starting in September 1952. Coverage not specified. No co-interventions.

Comparison group 1 (comparison groups 1 and 2 combined): Neighboring control area. No co-interventions.


OutcomesParasitaemia prevalence

No adverse event surveillance conducted

No adverse events reported


Notes


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNon-randomized controlled study

Allocation concealment (selection bias)High riskNon-randomized controlled study; selection of villages were made after initial survey. Communication facilities were taken into place to decide on the intervention.

Baseline imbalance (selection bias)High riskBaseline malaria characteristics were similar to comparison group 2 but not to comparison group 1.

Contamination protectionHigh riskIncidence of malaria was so high that every week large numbers of labourers were being repatriated to their own villages.

Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNot blinded, but unclear if this impacted outcomes

Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot blinded, but unclear if this impacted outcomes

Incomplete outcome data (attrition bias)
All outcomes
High riskNo description of intervention coverage

Selective reporting (reporting bias)Low riskEntire population surveyed

Other biasLow riskNo other bias detected

Song 2010 KHM

MethodsDates of study: 2003-2006

Location of study: Cambodia

Malaria endemicity (prevalence): 55.8% in children < 16 years; 52.3% in all ages [High]

Transmission season: Not described

Malaria species: P. falciparum, P. malariae, P. vivax

Vector species: Not described

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 3653

Intervention group 2 (mean): 2387


InterventionsIntervention group 1 (Kampong Speu, 17 villages, single round): MDA administered to all ages with artesunate 125 mg daily for two days, piperaquine 750 mg daily for two days and primaquine 9 mg every 10 days for six months starting in December 2003. Coverage not specified. No co-interventions.

Intervention group 2 (Kampot, nine villages, two rounds on days 0 and 42): MDA administered to all ages with artesunate 125 mg daily for two days and piperaquine 750 mg daily for two days given on days 0 and 42 and primaquine 9 mg every 10 days for six months starting in December 2003 . Coverage not specified. No co-interventions.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

Passive event surveillance conducted

Adverse events reported: No adverse reactions reported to village malaria volunteers.


NotesKampot data was not included in meta-analysis as the denominator of children for the outcome data was not provided.

The outcomes were assessed in a sub-sample of the treated population.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
High riskOne village missing data from one year

Selective reporting (reporting bias)High riskMonitoring was different for the different villages. Some villages had missing data.

Other biasLow riskNo other bias detected

van Dijk 1961 PNG

MethodsDates of study: 1960

Location of study: Papua New Guinea

Malaria endemicity (prevalence): 38.6% in children 2-9 years (1959 survey); 18% in all ages (1959 and 1960 surveys) [Moderate]

Transmission season: Not described

Malaria species:P. falciparum, P. malariae, P. vivax

Vector species: A. farauti

Study design: Uncontrolled before-and-after study

Evaluation design: Cross-sectional surveys


ParticipantsAge groups included: All ages

Sample size

Intervention group 1 (mean): 1250


InterventionsIntervention group 1: MDA administered to all persons with chloroquine (450 mg) once every four weeks for 11 rounds. Coverage 97.2% (range 93.1% to 100%). Co-intervention with mass treatment of filariasis with diethylcarbamazine.


OutcomesParasitaemia prevalence

Gametocytaemia prevalence

No adverse event surveillance conducted

No adverse events reported


NotesBefore MDA estimates include data from June 1959 and January 1960 surveys (Tables I and II). For intervention group 1, outcome estimates come from Table V.


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)High riskNo comparison group

Allocation concealment (selection bias)High riskNo comparison group

Baseline imbalance (selection bias)High riskNo comparison group

Contamination protectionHigh riskNo comparison group

Blinding of participants and personnel (performance bias)
All outcomes
High riskNo comparison group

Blinding of outcome assessment (detection bias)
All outcomes
High riskNo comparison group

Incomplete outcome data (attrition bias)
All outcomes
High riskNine positives were not included in the 0-1 month post-MDA survey; they were not present during the last distribution.

Selective reporting (reporting bias)Low riskAll pre-specified outcomes of interest have been reported

Other biasUnclear riskVisitors to the village were also treated with the group to which they were most closely related. Persons who stayed only a few days were not treated. However, it is unclear whether this introduced bias.

von Seidlein 2003 GMB

MethodsDates of study: 1999

Location of study: Gambia

Malaria endemicity: 42.9% in children ≤ 5 years [High]; describes 17-19% in all ages but this data was not from this study.

Transmission season: June to December

Malaria species: P. falciparum

Vector species: Not described

Study design: Cluster-randomized trial

Unit of randomization: Villages

Adjusted analyses for clustering: Yes

Adjustment method: Poisson regression model adjusting for population size

ICC: Not described

Number of clusters: 18 villages

Number of people: 3655

Average cluster size: 203

Feature: Matched villages

Evaluation design: Cross-sectional surveys, active surveillance and passive surveillance


ParticipantsAge groups included: Ages > 6 months old; non-pregnant

A total of 16,442 people, of which 14,017 people (85%) where treated (placebo or MDA) including the buffer zone

Sample size (of number evaluated)

Intervention group 1 (mean): 1969

Comparison group 1 (mean): 1686


InterventionsIntervention group 1: MDA administered to all non-pregnant persons aged > 6 months with sulfadoxine-pyrimethamine 1500 mg/75 mg and artesunate 200 mg once in June 1999. Coverage 89% in total population (90.8% in evaluated group). No co-interventions.

Comparison group 1: Placebo administered to all non-pregnant persons aged > 6 months once in June 1999. Coverage 89% in total population (89.6% in evaluated group). No co-interventions.


OutcomesParasitaemia prevalence

Parasitaemia incidence

Gametocytaemia prevalence

Anaemia prevalence (defined as hematocrit < 33%)

Mortality

Passive and active adverse event surveillance conducted

Adverse events reported (passive surveillance system): 1 episode of pruritus

Adverse events reported (active surveillance system): 25 of 75 individuals remembered one or more complaints within 2 days of taking the drug including dizziness (13), fever (6), diarrhoea (5), vomiting (5) and itching (4).


NotesIncidence, gametocyte prevalence, anaemia prevalence and mortality reported for children only


Risk of bias

BiasAuthors' judgementSupport for judgement

Random sequence generation (selection bias)Low riskWhile the study is a cluster-randomized, double blind, placebo-controlled trial, the method of randomization is not described. Author correspondence revealed that randomization was computer generated.

Allocation concealment (selection bias)Low riskDrugs allocated to each of the 18 study villages were delivered to the study site in identical containers. One nurse was aware of the identity of the drugs, administered the drugs in the study villages and then left the study area.

Baseline imbalance (selection bias)Low riskIntervention and control villages did not differ appreciably in the demographic of malaria transmission characteristics.

Contamination protectionLow riskAll inhabitants of the non-randomized controlled villages in the study area were treated, to minimize possible dilution of the effect of the intervention.

Blinding of participants and personnel (performance bias)
All outcomes
Low riskCluster-randomized, double blind, placebo-controlled trial; neither study personnel nor the study population were aware of which villages received placebo.

Blinding of outcome assessment (detection bias)
All outcomes
Low riskCluster-randomized, double blind, placebo-controlled trial

Incomplete outcome data (attrition bias)
All outcomes
Low riskAll children in the surveillance villages were visited weekly; all 18 study villages that were randomized were analyzed.

Selective reporting (reporting bias)Low riskAll primary and secondary endpoints reported

Other biasLow riskNo other bias detected

 
Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion

Abraham 1944Inadequate treatment dose

Afridi 1959Inadequate treatment dose

Ahorlu 2009Treatment not administered to entire population; intermittent preventive treatment for children (IPTc) study

Ahorlu 2011Treatment not administered to entire population; intermittent preventive treatment for children (IPTc) study

Aikins 1993Inadequate treatment dose; knowledge, attitudes, and prevention component of an individually randomized study

Alicata 1955Inadequate treatment dose; individually randomized study

Aliev 2000Inadequate treatment dose

Aliev 2001Inadequate treatment dose

Allen 1990Inadequate treatment dose

Alonso 1993aInadequate treatment dose; individually randomized study

Alonso 1993bInadequate treatment dose; individually randomized study

Alving 1952Individually randomized study; study participants did not remain in endemic area

Amangel'diev 2001Inadequate treatment dose; testing conducted prior to treatment; insufficient information on drug administration

Annual Report 1932Inadequate treatment dose

Archambeault 1954Study participants did not remain in endemic area

Archibald 1956Individually randomized study

Babione 1966Insufficient information on drug administration

Banerjea 1949Inadequate treatment dose

Barber 1932Inadequate treatment dose

Barger 2009Individually randomized study

Baukapur 1984Insufficient information on drug administration

Berberian 1948Testing conducted prior to treatment

Berny 1936Inadequate treatment dose

Bloch 1982Insufficient information on drug administration

Bojang 2009Insufficient information on outcomes reported

Bojang 2010Individually randomized study

Bojang 2011Insufficient information on outcomes reported

Boulanger 2009Individually randomized study

Boulanger 2010Individually randomized study

Brink 1958Inadequate treatment dose

Butler 1943Insufficient information on drug administration

Canet 1936Inadequate treatment dose

Canet 1939Insufficient information on outcomes reported

Canet 1949Insufficient information on drug administration; insufficient information on outcomes reported (no outcome of interest reported)

Canet 1952Inadequate treatment dose

Canet 1953Inadequate treatment dose

Capponi 1953Inadequate treatment dose

Celli 1914Insufficient information on drug administration

Charles 1958Individually randomized study; testing conducted prior to treatment

Charles 1960Inadequate treatment dose

Charles 1962Inadequate treatment dose

Chaudhuri 1950Inadequate treatment dose

Chen 1999Insufficient information on outcomes reported; treatment not administered to entire population

Cisse 2006Individually randomized study

Cisse 2009Treatment not administered to entire population; intermittent preventive treatment for children (IPTc) study

Ciuca 1937Mixed curative and prophylactic dosing

Clark 1942Testing conducted prior to treatment

Clarke 2008Treatment not administered to entire population; intermittent preventive treatment for children (IPTc) study

Clyde 1958Insufficient information on drug administration

Clyde 1961aInsufficient information on outcomes reported

Clyde 1961bInsufficient information on outcomes reported

Clyde 1962Inadequate treatment dose

Coutinho 1962Inadequate treatment dose

D'Anfreville 1930Insufficient information on drug administration; insufficient information on outcomes reported (no outcome of interest reported)

Danquah 2009Individually randomized study

Dapeng 1996Insufficient information on drug administration; insufficient information on outcomes reported

de Mello 1938Inadequate treatment dose

Decourt 1935Inadequate treatment dose; individually randomized study

Decourt 1936Inadequate treatment dose

Delmont 1981Inadequate treatment dose; individually randomized study

Desowitz 1987Insufficient information on drug administration

Diallo 1977Inadequate treatment dose

Diallo 1983Treatment not administered to entire population; intermittent preventive treatment in children (IPTc) study

Dicko 2008Individually randomized study; testing conducted prior to treatment

Dicko 2011Individually randomized study

Dixon 1950Inadeqaute treatment dose

Doi 1989Individually randomized study; testing conducted prior to treatment

Dola 1974Inadequate treatment dose

Doucet 1947Inadequate treatment dose

Downs 1946Study participants did not remain in endemic area

Dupoux 1937Insufficient information on outcomes reported

Dupoux 1939Inadequate treatment dose

Edeson 1957Inadequate treatment dose

Farinaud 1934Insufficient information on drug administration

Farinaud 1950Inadequate treatment dose

Gaud 1949Inadequate treatment dose

Gilroy 1952Inadequate treatment dose

Gomez Mendoza 1960Insufficient information on outcomes reported

Gribben 1933Inadequate treatment dose

Gruer 1962Insufficient information on drug administration; insufficient information on outcomes of interest

Gunther 1951Inadequate treatment dose

Gunther 1952Mixed curative and prophylactic dosing

Gusmao 1970Inadequate treatment dose; individually randomized study

Han 2006Inadequate treatment dose

Harwin 1973Individually randomized study

Henderson 1934Mixed curative and prophylactic dosing

Ho 1965Insufficient information on outcomes reported (no outcome of interest reported)

Houel 1954bTreatment not administered to entire population (children only)

Huehne 1971Post-only outcomes reported

Janssens 1950Inadequate treatment dose

Joncour 1956Inadequate treatment dose

Kaneko 2010Insufficient information on drug administration; insufficient information on outcomes reported.

Karimov 2008Inadequate treatment dose

Kingsbury 1931Inadequate treatment dose

Klopfer 1949Inadequate treatment dose

Komp 1935Testing conducted prior to treatment

Konate 2011Individually randomized study

Kweku 2008Individually randomized study

Kweku 2009Insufficient information on outcomes reported; comparison of delivery strategies; treatment not administered to entire population (both arms included intermittent preventive treatment in children (IPTc))

Lacroix 1952Inadequate treatment dose

Lahon 1960Inadequate treatment dose

Laing 1970Testing conducted prior to treatment

Laing 1984Inadequate treatment dose

Lakshmanacharyulu 1968Insufficient information on drug administration

Levenson 1943Mixed curative and prophylactic dosing

Liljander 2010Individually randomized study

Lui 1986Mixed curative and prophylactic dosing

Lysenko 1960Mixed curative and prophylactic dosing

MacCormack 1983Inadequate treatment dose

Mackerras 1954Inadequate treatment dose

Maiga 2009Individually randomized study

Malaria_Army 1934Inadequate treatment dose; insufficient information on outcomes reported

Mason 1973Insufficient information on drug administration

Mason 1977Insufficient information on drug administration

Mastbaum 1957aInadequate treatment dose

Mastbaum 1957bInadequate treatment dose

McGregor 1966Individually randomized study; testing conducted prior to treatment

Melik-Adamian 1938Testing conducted prior to treatment

Mendez Galvan 1984Insufficient information on drug administration; insufficient information on outcomes reported

Mercier 1953Inadequate treatment dose

Merle 1955Inadequate treatment dose; treatment not administered to entire population (eg intermittent preventive treatment for children (IPTc))

Mezincesco 1935Inadequate treatment dose

Miller 1955Inadequate treatment dose; individually randomized study; treatment not administered to entire population

Monteny 1960Inadequate treatment dose

Mühlens 1913Insufficient information on drug administration; insufficient information on outcomes of interest

Nakibuuka 2009Individually randomized study; testing conducted prior to treatment

Nankabirwa 2010Individually randomized study

Nave 1973Insufficient information on outcomes reported

Norman 1952Inadequate treatment dose; insufficient information on outcomes of interest

Ntab 2007Individually randomized study; insufficient information on outcomes reported

Omer 1978Inadequate treatment dose

Onori 1972Inadequate treatment dose

Ossi 1967Insufficient information on outcomes reported

Ouedraogo 2010Individually randomized study

Parrot 1937Inadequate treatment dose

Parrot 1943Inadequate treatment dose

Parrot 1944Inadequate treatment dose

Parrot 1946Inadequate treatment dose

Peters 1962Inadequate treatment dose

Phillips 1954Inadequate treatment dose

Pikul 1934Insufficient information on outcomes reported

Pribadi 1986Inadequate treatment dose

Prokopenko 1945Inadequate treatment dose

Rachou 1965Inadequate treatment dose

Rafi 1951Inadequate treatment dose

Ray 1948Inadequate treatment dose

Robin 1946Testing conducted prior to treatment

Rodríguez 1994Testing conducted prior to treatment

Rohner 2010Individually randomized study

Saarinen 1987Mixed curative and prophylactic dosing

Salako 1990Individually randomized study; testing conducted prior to treatment

Salihu 2000Inadequate treatment dose

Santos 1993Inadequate treatment dose

Schliessmann 1973Insufficient information on outcomes reported

Schneider 1948aInadequate treatment dose

Schneider 1948bInadequate treatment dose

Schneider 1958Inadequate treatment dose

Schneider 1962Individually randomized study; treatment not administered to entire population (eg intermittent preventive treatment for children (IPTc))

Seckinger 1935Inadequate treatment dose

Sehgal 1968Insufficient information on drug administration

Sergent 1913Inadequate treatment dose; insufficient information on outcomes reported

Sesay 2011Individually randomized study

Shanks 1992Inadequate treatment dose; individually randomized study; testing conducted prior to treatment

Shanks 1993Individually randomized study; study participants did not remain in endemic area

Shanks 1995aInadequate treatment dose; study participants did not remain in endemic area

Shanks 1995bStudy participants did not remain in endemic area

Sheinker 1945Testing conducted prior to treatment

Singh 1968Insufficient information on outcomes reported

Snowden 2006Insufficient information on drug administration; insufficient information on outcomes reported

Sokhna 2008Individually randomized study

Sorel 1913Insufficient information on drug administration

Srivastava 1950Inadequate treatment dose

Strangeways-Dixon 1950Inadequate treatment dose

Strickland 1986Testing conducted prior to treatment

Swellengrebel 1931Inadequate treatment dose

Tagbor 2011Treatment not administered to entire population; intermittent preventive treatment for children (IPTc) study

Tine 2011Treatment not administered to entire population; intermittent preventive treatment for children (IPTc) study

Turner 1977Insufficient information on outcomes reported

Usenbaev 2006Insufficient information on drug administration

Usenbaev 2008Insufficient information on drug administration

Van Dijk 1958Inadequate treatment dose

Van Goor 1950Inadequate treatment dose

Verhoef 2002Individually randomized study

Villegas 2010Testing conducted prior to treatment

Wallace 1936Mixed curative and prophylactic dosing

Wallace 1954Insufficient information on drug administration

Watkins 1987Individually randomized; mixed curative and prophylactic dosing

White 1934Inadequate treatment dose

White 1937Mixed curative and prophylactic dosing

Winter 1934Insufficient information on outcomes reported

Wone 1967Inadequate treatment dose

Yip 1998Insufficient information on outcomes reported

 
Comparison 1. MDA versus no MDA in areas of low endemicity (Stratified by study design)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Parasitaemia Prevalence: Cluster-randomized trials1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    1.1 At baseline
1496Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

    1.2 <1 month post MDA
1484Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

    1.3 1-3 months post MDA
1794Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

    1.4 4-6 months post MDA
1660Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

 2 Parasitaemia Prevalence: Uncontrolled before-and-after studies1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    2.1 <1 month post MDA
13039Risk Ratio (M-H, Random, 95% CI)0.27 [0.14, 0.50]

    2.2 >12 months post MDA
13509Risk Ratio (M-H, Random, 95% CI)0.02 [0.00, 0.12]

 3 Gametocytaemia Prevalence: Cluster randomized trials1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    3.1 At baseline
1496Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

    3.2 < 1 month post MDA
1484Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

    3.3 1-3 months post MDA
1794Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

    3.4 4-6 months post MDA
1660Risk Ratio (M-H, Random, 95% CI)0.0 [0.0, 0.0]

 
Comparison 2. MDA versus no MDA in areas of moderate endemicity (Stratified by study design)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Parasitaemia Prevalence: Non-randomized controlled studies4Risk Ratio (M-H, Random, 95% CI)Subtotals only

    1.1 At baseline
43123Risk Ratio (M-H, Random, 95% CI)0.73 [0.43, 1.24]

    1.2 During MDA
147014Risk Ratio (M-H, Random, 95% CI)0.27 [0.25, 0.28]

    1.3 < 1 month post MDA
31934Risk Ratio (M-H, Random, 95% CI)0.03 [0.01, 0.08]

    1.4 1-3 months post MDA
21557Risk Ratio (M-H, Random, 95% CI)0.15 [0.10, 0.23]

    1.5 4-6 months post MDA
21610Risk Ratio (M-H, Random, 95% CI)0.18 [0.10, 0.33]

    1.6 7-12 months post MDA
1600Risk Ratio (M-H, Random, 95% CI)0.19 [0.11, 0.33]

 2 Parasitaemia Prevalence: Uncontrolled before-and-after studies7Risk Ratio (M-H, Random, 95% CI)Subtotals only

    2.1 During MDA
27965Risk Ratio (M-H, Random, 95% CI)0.17 [0.02, 1.47]

    2.2 <1 month post MDA
33096Risk Ratio (M-H, Random, 95% CI)0.29 [0.17, 0.48]

    2.3 1-3 months post MDA
47925Risk Ratio (M-H, Random, 95% CI)0.16 [0.08, 0.31]

    2.4 4-6 months post MDA
23797Risk Ratio (M-H, Random, 95% CI)1.75 [0.41, 7.41]

 3 Gametocytaemia Prevalence: Non-randomized controlled studies2Risk Ratio (M-H, Random, 95% CI)Subtotals only

    3.1 At baseline
21622Risk Ratio (M-H, Random, 95% CI)1.40 [0.76, 2.57]

    3.2 During MDA
147014Risk Ratio (M-H, Random, 95% CI)0.48 [0.42, 0.54]

    3.3 <1 month post MDA
1433Risk Ratio (M-H, Random, 95% CI)0.28 [0.10, 0.82]

    3.4 1-3 months post MDA
1357Risk Ratio (M-H, Random, 95% CI)0.17 [0.03, 0.86]

    3.5 4-6 months post MDA
1410Risk Ratio (M-H, Random, 95% CI)0.52 [0.24, 1.11]

 4 Gametocytaemia Prevalence: Uncontrolled before-and-after studies3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    4.1 <1 month post MDA
33096Risk Ratio (M-H, Random, 95% CI)0.47 [0.25, 0.87]

    4.2 1-3 months post MDA
1294Risk Ratio (M-H, Random, 95% CI)0.36 [0.12, 1.12]

    4.3 4-6 months post MDA
1204Risk Ratio (M-H, Random, 95% CI)0.35 [0.12, 1.01]

 
Comparison 3. MDA versus no MDA in areas of high endemicity (Stratified by study design)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Parasitaemia Prevalence: Cluster-randomized trials1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    1.1 At baseline
11376Risk Ratio (M-H, Random, 95% CI)0.97 [0.86, 1.10]

    1.2 1-3 months post MDA
11800Risk Ratio (M-H, Random, 95% CI)0.82 [0.67, 1.01]

    1.3 4-6 months post MDA
11089Risk Ratio (M-H, Random, 95% CI)1.16 [0.93, 1.44]

 2 Parasitaemia Prevalence: Non-randomized controlled studies3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    2.1 At baseline
39395Risk Ratio (M-H, Random, 95% CI)0.84 [0.70, 1.00]

    2.2 During MDA
312561Risk Ratio (M-H, Random, 95% CI)0.17 [0.11, 0.27]

    2.3 1-3 months post MDA
27197Risk Ratio (M-H, Random, 95% CI)0.52 [0.33, 0.81]

 3 Parasitaemia Prevalence: Uncontrolled before-and-after studies7Risk Ratio (M-H, Random, 95% CI)Subtotals only

    3.1 During MDA
22011Risk Ratio (M-H, Random, 95% CI)0.10 [0.03, 0.34]

    3.2 <1 month post MDA
43863Risk Ratio (M-H, Random, 95% CI)0.37 [0.28, 0.49]

    3.3 1-3 months post MDA
45132Risk Ratio (M-H, Random, 95% CI)0.35 [0.15, 0.84]

    3.4 4-6 months post MDA
32979Risk Ratio (M-H, Random, 95% CI)0.41 [0.24, 0.72]

    3.5 7-12 months post MDA
175Risk Ratio (M-H, Random, 95% CI)0.72 [0.43, 1.20]

    3.6 >12 months post MDA
12375Risk Ratio (M-H, Random, 95% CI)0.10 [0.07, 0.12]

 4 Parasitaemia Incidence: Cluster-randomized trials1Rate Ratio (Random, 95% CI)0.84 [0.53, 1.32]

    4.1 < 1 month post MDA
1Rate Ratio (Random, 95% CI)0.41 [0.23, 0.74]

    4.2 1-3 months post MDA
1Rate Ratio (Random, 95% CI)1.03 [0.75, 1.41]

    4.3 4-6 months post MDA
1Rate Ratio (Random, 95% CI)1.11 [0.84, 1.45]

 5 Gametocytaemia Prevalence: Cluster-randomized trials1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    5.1 At baseline
11376Risk Ratio (M-H, Random, 95% CI)0.66 [0.33, 1.29]

    5.2 4-6 months post MDA
11414Risk Ratio (M-H, Random, 95% CI)1.07 [0.62, 1.85]

 6 Gametocytaemia Prevalence: Non-randomized controlled studies3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    6.1 At baseline
39395Risk Ratio (M-H, Random, 95% CI)0.72 [0.55, 0.95]

    6.2 During MDA
312561Risk Ratio (M-H, Random, 95% CI)0.17 [0.10, 0.28]

    6.3 1-3 months post MDA
27197Risk Ratio (M-H, Random, 95% CI)0.55 [0.28, 1.07]

 7 Gametocytaemia Prevalence: Uncontrolled before-and-after studies5Risk Ratio (M-H, Random, 95% CI)Subtotals only

    7.1 During MDA
22011Risk Ratio (M-H, Random, 95% CI)0.35 [0.09, 1.40]

    7.2 <1 month post MDA
32582Risk Ratio (M-H, Random, 95% CI)0.38 [0.13, 1.08]

    7.3 1-3 months post MDA
21199Risk Ratio (M-H, Random, 95% CI)1.14 [0.64, 2.01]

    7.4 4-6 months post MDA
22789Risk Ratio (M-H, Random, 95% CI)0.35 [0.10, 1.28]

    7.5 7-12 months post MDA
175Risk Ratio (M-H, Random, 95% CI)0.86 [0.41, 1.79]

    7.6 >12 months post MDA
12269Risk Ratio (M-H, Random, 95% CI)0.09 [0.05, 0.15]

 8 Anaemia Prevalence: Cluster-randomized trials1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    8.1 4-6 months post MDA
11414Risk Ratio (M-H, Random, 95% CI)0.84 [0.75, 0.93]

 9 Mortality: Cluster-randomized trials1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    9.1 4-6 months post MDA
13655Risk Ratio (M-H, Random, 95% CI)1.43 [0.34, 5.96]

 
Comparison 4. MDA + vector control versus no intervention in areas of moderate endemicity (Stratified by study design)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Parasitaemia Prevalence: Non-randomized controlled studies1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    1.1 At baseline
11080Risk Ratio (M-H, Random, 95% CI)2.09 [1.48, 2.98]

    1.2 >12 months post MDA
11331Risk Ratio (M-H, Random, 95% CI)0.10 [0.05, 0.20]

 2 Parasitaemia Prevalence: Uncontrolled before-and-after studies4Risk Ratio (M-H, Random, 95% CI)Subtotals only

    2.1 During MDA
22336Risk Ratio (M-H, Random, 95% CI)0.12 [0.02, 0.62]

    2.2 <1 month post MDA
35006Risk Ratio (M-H, Random, 95% CI)0.06 [0.01, 0.33]

    2.3 1-3 months post MDA
34724Risk Ratio (M-H, Random, 95% CI)0.14 [0.04, 0.57]

    2.4 4-6 months post MDA
1939Risk Ratio (M-H, Random, 95% CI)0.57 [0.39, 0.85]

    2.5 >12 months post MDA
11758Risk Ratio (M-H, Random, 95% CI)0.00 [5.43, 0.03]

 3 Gametocytaemia Prevalence: Uncontrolled before-and-after studies2Risk Ratio (M-H, Random, 95% CI)Subtotals only

    3.1 During MDA
24425Risk Ratio (M-H, Random, 95% CI)0.13 [0.06, 0.27]

    3.2 < 1 month post MDA
11907Risk Ratio (M-H, Random, 95% CI)0.01 [6.21, 0.16]

    3.3 1-3 months post MDA
11941Risk Ratio (M-H, Random, 95% CI)0.22 [0.11, 0.41]

 
Comparison 5. MDA + vector control versus no intervention in areas of high endemicity (Stratified by study design)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Parasitaemia Prevalence: Non-randomized controlled studies3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    1.1 At baseline
38042Risk Ratio (M-H, Random, 95% CI)0.56 [0.37, 0.84]

    1.2 During MDA
39493Risk Ratio (M-H, Random, 95% CI)0.10 [0.06, 0.16]

    1.3 1-3 months post MDA
24455Risk Ratio (M-H, Random, 95% CI)0.12 [0.06, 0.23]

    1.4 7-12 months post MDA
13154Risk Ratio (M-H, Random, 95% CI)0.60 [0.55, 0.67]

    1.5 >12 months post MDA
13261Risk Ratio (M-H, Random, 95% CI)0.77 [0.70, 0.84]

 2 Parasitaemia Prevalence: Uncontrolled before-and-after studies2Risk Ratio (M-H, Random, 95% CI)Subtotals only

    2.1 During MDA
25437Risk Ratio (M-H, Random, 95% CI)0.17 [0.09, 0.31]

    2.2 1-3 months post MDA
25440Risk Ratio (M-H, Random, 95% CI)0.13 [0.01, 2.51]

    2.3 4-6 months post MDA
1415Risk Ratio (M-H, Random, 95% CI)0.83 [0.66, 1.04]

    2.4 7-12 months post MDA
1412Risk Ratio (M-H, Random, 95% CI)0.93 [0.75, 1.16]

 3 Gametocytaemia Prevalence: Non-randomized controlled studies3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    3.1 At baseline
38042Risk Ratio (M-H, Random, 95% CI)0.53 [0.31, 0.90]

    3.2 During MDA
39493Risk Ratio (M-H, Random, 95% CI)0.08 [0.03, 0.20]

    3.3 1-3 months post MDA
24455Risk Ratio (M-H, Random, 95% CI)0.08 [0.05, 0.14]

    3.4 7-12 months post MDA
13154Risk Ratio (M-H, Random, 95% CI)0.87 [0.73, 1.05]

    3.5 > 12 months post MDA
13261Risk Ratio (M-H, Random, 95% CI)0.96 [0.81, 1.14]

 4 Gametocytaemia Prevalence: Uncontrolled before-and-after studies1Risk Ratio (M-H, Random, 95% CI)Subtotals only

    4.1 During MDA
1437Risk Ratio (M-H, Random, 95% CI)0.29 [0.17, 0.50]

    4.2 1-3 months post MDA
1440Risk Ratio (M-H, Random, 95% CI)0.52 [0.34, 0.80]

    4.3 4-6 months post MDA
1415Risk Ratio (M-H, Random, 95% CI)0.76 [0.52, 1.12]

    4.4 7-12 months post MDA
1412Risk Ratio (M-H, Random, 95% CI)0.93 [0.65, 1.33]

 
Comparison 6. Parasitaemia Incidence studies

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 MDA versus no MDA: Uncontrolled before-and-after studies4Rate Ratio (Random, 95% CI)Subtotals only

    1.1 During MDA
3Rate Ratio (Random, 95% CI)0.29 [0.07, 1.14]

    1.2 < 1 month post MDA
4Rate Ratio (Random, 95% CI)0.21 [0.05, 0.84]

    1.3 1-3 months post MDA
4Rate Ratio (Random, 95% CI)0.61 [0.26, 1.40]

    1.4 4-6 months post MDA
1Rate Ratio (Random, 95% CI)0.65 [0.41, 1.02]

    1.5 7-12 months post MDA
1Rate Ratio (Random, 95% CI)0.15 [0.07, 0.34]

    1.6 >12 months post MDA
1Rate Ratio (Random, 95% CI)0.48 [0.42, 0.55]

 2 MDA + vector control versus no MDA: Uncontrolled before-and-after studies2Rate Ratio (Random, 95% CI)Subtotals only

    2.1 During MDA
2Rate Ratio (Random, 95% CI)0.92 [0.49, 1.75]

    2.2 < 1 month post MDA
2Rate Ratio (Random, 95% CI)0.04 [8.50, 1.54]

    2.3 1-3 months post MDA
2Rate Ratio (Random, 95% CI)0.08 [0.01, 0.98]

    2.4 4-6 months post MDA
2Rate Ratio (Random, 95% CI)0.11 [0.01, 1.97]

    2.5 7-12 months post MDA
2Rate Ratio (Random, 95% CI)0.16 [0.01, 3.10]

    2.6 > 12 months post MDA
1Rate Ratio (Random, 95% CI)0.04 [0.03, 0.07]

 
Comparison 7. MDA versus no MDA in areas of moderate and high endemicity (Stratified by study design; subgrouped by 8-aminoquinoline)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Parasitaemia Prevalence during MDA4Risk Ratio (M-H, Random, 95% CI)Subtotals only

    1.1 Non-randomized controlled studies - with 8-aminoquinoline
26634Risk Ratio (M-H, Random, 95% CI)0.20 [0.12, 0.32]

    1.2 Non-randomized controlled studies - without 8-aminoquinoline
252941Risk Ratio (M-H, Random, 95% CI)0.16 [0.08, 0.31]

 2 Parasitaemia Prevalence 1-3 months post MDA4Risk Ratio (M-H, Random, 95% CI)Subtotals only

    2.1 Non-randomized controlled studies - with 8-aminoquinoline
27197Risk Ratio (M-H, Random, 95% CI)0.52 [0.33, 0.81]

    2.2 Non-randomized controlled studies - without 8-aminoquinoline
21557Risk Ratio (M-H, Random, 95% CI)0.15 [0.10, 0.23]

 3 Parasitaemia Prevalence during MDA4Risk Ratio (M-H, Random, 95% CI)Subtotals only

    3.1 Uncontrolled before-and-after studies - with 8-aminoquinoline
12965Risk Ratio (M-H, Random, 95% CI)0.06 [0.03, 0.10]

    3.2 Uncontrolled before-and-after studies - without 8-aminoquinoline
37011Risk Ratio (M-H, Random, 95% CI)0.17 [0.06, 0.51]

 4 Parasitaemia Prevalence <1 month post MDA7Risk Ratio (M-H, Random, 95% CI)Subtotals only

    4.1 Uncontrolled before-and-after studies - with 8-aminoquinoline
32650Risk Ratio (M-H, Random, 95% CI)0.42 [0.29, 0.61]

    4.2 Uncontrolled before-and-after studies - without 8-aminoquinoline
44309Risk Ratio (M-H, Random, 95% CI)0.29 [0.22, 0.38]

 5 Parasitaemia Prevalence 1-3 months post MDA7Risk Ratio (M-H, Fixed, 95% CI)Subtotals only

    5.1 Uncontrolled before-and-after studies - with 8-aminoquinoline
198Risk Ratio (M-H, Fixed, 95% CI)0.65 [0.41, 1.01]

    5.2 Uncontrolled before-and-after studies - without 8-aminoquinoline
612959Risk Ratio (M-H, Fixed, 95% CI)0.32 [0.29, 0.34]

 6 Parasitaemia Prevalence 4-6 months post MDA5Risk Ratio (M-H, Random, 95% CI)Subtotals only

    6.1 Uncontrolled before-and-after studies - with 8-aminoquinoline
32979Risk Ratio (M-H, Random, 95% CI)0.41 [0.24, 0.72]

    6.2 Uncontrolled before-and-after studies - without 8-aminoquinoline
23797Risk Ratio (M-H, Random, 95% CI)1.75 [0.41, 7.41]

 
Comparison 8. MDA versus no MDA for all endemicity levels (Stratified by study design; subgrouped by plasmodium species)

Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size

 1 Parasitaemia Prevalence at baseline2Risk Ratio (M-H, Random, 95% CI)Subtotals only

    1.1 Non-randomized controlled studies - falciparum
21537Risk Ratio (M-H, Random, 95% CI)1.34 [1.03, 1.74]

    1.2 Non-randomized controlled studies - vivax
21537Risk Ratio (M-H, Random, 95% CI)3.84 [1.33, 11.04]

 2 Parasitaemia Prevalence during MDA2Risk Ratio (M-H, Random, 95% CI)Subtotals only

    2.1 Uncontrolled before-and-after studies - falciparum
25561Risk Ratio (M-H, Random, 95% CI)0.40 [0.08, 1.97]

    2.2 Uncontrolled before-and-after studies - vivax
25561Risk Ratio (M-H, Random, 95% CI)0.60 [0.40, 0.90]

 3 Parasitaemia Prevalence <1 month post MDA5Risk Ratio (M-H, Random, 95% CI)Subtotals only

    3.1 Non-randomized controlled studies - falciparum
1433Risk Ratio (M-H, Random, 95% CI)0.02 [0.00, 0.08]

    3.2 Non-randomized controlled studies - vivax
1433Risk Ratio (M-H, Random, 95% CI)0.05 [0.00, 0.82]

    3.3 Uncontrolled before-and-after studies - falciparum
47367Risk Ratio (M-H, Random, 95% CI)0.22 [0.18, 0.29]

    3.4 Uncontrolled before-and-after studies - vivax
47367Risk Ratio (M-H, Random, 95% CI)0.50 [0.41, 0.61]

 4 Parasitaemia Prevalence 1-3 months post MDA3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    4.1 Non-randomized controlled studies - falciparum
1357Risk Ratio (M-H, Random, 95% CI)0.03 [0.01, 0.12]

    4.2 Non-randomized controlled studies - vivax
1357Risk Ratio (M-H, Random, 95% CI)1.37 [0.46, 4.11]

    4.3 Uncontrolled before-and-after studies - falciparum
25754Risk Ratio (M-H, Random, 95% CI)0.22 [0.09, 0.51]

    4.4 Uncontrolled before-and-after studies - vivax
25754Risk Ratio (M-H, Random, 95% CI)0.49 [0.32, 0.76]

 5 Parasitaemia Prevalence 4-6 months post MDA3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    5.1 Non-randomized controlled studies - falciparum
1410Risk Ratio (M-H, Random, 95% CI)0.22 [0.14, 0.33]

    5.2 Non-randomized controlled studies - vivax
1410Risk Ratio (M-H, Random, 95% CI)0.81 [0.31, 2.08]

    5.3 Uncontrolled before-and-after studies - falciparum
23642Risk Ratio (M-H, Random, 95% CI)0.40 [0.13, 1.23]

    5.4 Uncontrolled before-and-after studies - vivax
23642Risk Ratio (M-H, Random, 95% CI)0.31 [0.24, 0.39]

 6 Parasitaemia Prevalence >12 months post MDA3Risk Ratio (M-H, Random, 95% CI)Subtotals only

    6.1 Non-randomized controlled studies - falciparum
11331Risk Ratio (M-H, Random, 95% CI)0.01 [4.99, 0.13]

    6.2 Non-randomized controlled studies - vivax
11331Risk Ratio (M-H, Random, 95% CI)0.36 [0.15, 0.86]

    6.3 Uncontrolled before-and-after studies - falciparum
25884Risk Ratio (M-H, Random, 95% CI)0.06 [0.04, 0.09]

    6.4 Uncontrolled before-and-after studies - vivax
25884Risk Ratio (M-H, Random, 95% CI)0.17 [0.12, 0.24]

 
Table 1. Overview of studies conducted in areas of low endemicity

Study IDDesignCountryYearEndemicityMDA groupControl group/

Baseline

Drug (dose)IntervalNo. of roundsPopulation
targeted
CoverageCo-intervention

Shekalaghe 2011CRTTanzania20080%*AS (4 mg/kg/day for 3 days) +SP (25 mg/1.25 mg on day 1) +PQ (0.75 mg on day 3)-1111093%Background ITN usePlacebo + Background ITN use

Malaria_Taiwan 1991BASTaiwan19553-4%*CQ (12 mg/kg)-11520NDIRSIRS

 CRT = Cluster-randomized trial; BAS = Uncontrolled before-and-after study; AS = Artesunate; SP = Sulfadoxine-Pyrimethamine; PQ = Primaquine; CQ = Chloroquine; ND = Not described; IRS = Indoor Residual Spraying.
*In all ages
 
Table 2. Overview of studies conducted in areas of moderate endemicity

Study IDDesignCountryYearEndemicityMDA groupControl group/baseline

Drug (dose)IntervalNo. of roundsPopulation
targeted
CoverageCo-intervention

Najera 1973N-RCSNigeria1966-6829%*CQ (450 mg) + Pyr (45 mg)2 months1152,00078-92%IRSIRS alone

Singh 1953N-RCSIndia1952-5322%*AQ (600 mg)2 weeks5125NDNoneNo intervention

Jones 1958N-RCSKenya1952-5334%Pyr (100 mg)6 months33721-4500NDNoneNo intervention

Roberts 1964N-RCSKenya195328%*Pyr (50 mg)-1101,00095%NoneNo intervention

N-RCSKenya195422%*Pyr (50 mg)-199,00095%NoneNo intervention

Archibald 1960BASNigeria195829%CQ (600 mg) + Pyr (25 mg)1 month510,000NDIRSIRS

Cavalie 1962BASCameroon1960-6120%CQ (600 mg) + Pyr (50 mg)4 months222,50076-92%IRSIRS

Houel 1954BASMorocco195314%Pyr (100 mg)-19999NDIRSIRS

Metselaar 1961BASNew Guinea1958-5913-21%CQ (450 mg) +Pyr (50 mg)1 week6250090%IRSIRS

Jones 1954BASKenya1952-5335%Pyr (100 mg)6 months33721NDNone-

van Dijk 1961BASPapua New Guinea196039%CQ (450 mg)4 weeks11125097%None-

Comer 1971BASPanama1965-6817%*Pyr (50 mg / 75 mg) + PQ (40 mg)2 weeks49170961-87%None-

 N-RCS = Non-randomized controlled study; BAS = Uncontrolled before-and-after study; AQ = Amodiaquine; Pyr = Pyrimethamine; CQ = Chloroquine; PQ = Primaquine; ND = Not described; IRS = Indoor Residual Spraying.
*In all ages
Amongst children only
 
Table 3. Overview of studies conducted in areas of high endemicity

Study IDDesignCountryYearEndemicityMDA groupControl group

Drug (dose)IntervalNo. of roundsPopulation
targeted
CoverageCo-intervention

Von Seidlein 2003CRTGambia199943%AS (4 mg/kg/day for 3 days) +SP (25 mg/1.25 mg on day 1)-1196989%NonePlacebo

Molineaux 1980N-RCSNigeria1970-7546%*SP (500 mg/25 mg)10 weeks9‡14,12985%IRSIRS alone

SP (500 mg/25 mg)2-10 weeks23‡181085%IRSIRS alone

Escudie 1962N-RCSBurkina Faso1960-6156.1%CQ (600 mg)/AQ (600 mg) +PQ (15 mg)1 month8189075-92%NoneNo intervention

CQ (600 mg)/AQ (600 mg) +PQ (15 mg)2 weeks15256084-97%NoneNo intervention

CQ (600 mg)/AQ (600 mg) +PQ (15 mg)1 month8540081-92%IRSIRS alone-

CQ (600 mg)/AQ (600 mg) +PQ (15 mg)2 weeks15349082-94%IRSIRS alone-

Schneider 1961N-RCSBurkina Faso1960-6159%CQ (600 mg)/AQ (600 mg) +PQ (15 mg)2 weeks15250090%NoneNo intervention

Archibald 1960BASNigeria1957-5964%CQ (600 mg) +Pyr (25 mg)6 months41300NDIRSIRS-

Cavalie 1962BASCameroon1960-6165%*CQ (600 mg) +Pyr (50 mg)-17000100%IRSIRS

Gaud 1953BASMorocco195242%*AQ (600 mg)-13000NDNone-

Ricosse 1959BASBurkina Faso1958-5956%Pyr (50 mg)2 weeks8300082-91%None-

Song 2010BASCambodia2003-0656%AS (125 mg/day for 2 days) + PIP (750 mg/day for 2 days) + PQ (9 mg every 10 days)-13653NDNone-

Hii 1987BASMalaysia1984-8556%SP (1500 mg / 75 mg) + PQ (30 mg)-114876%None-

Kligler 1931BASPalestine193067%Plas (30 mg) + Q (900 mg) twice daily for 5 days3 weeks395379%None-

 CRT= Cluster-randomized trial; N-RCS = Non-randomized controlled study; BAS = Uncontrolled before-and-after study; AS = Artesunate; SP = Sulfadoxine (or sulfalene)-Pyrimethamine; Pyr = Pyrimethamine; CQ = Chloroquine; AQ = Amodiaquine; PQ = Primaquine; Pip = Piperaquine; Plas = Plasmochin; Q = Quinine; ND = Not described; IRS = Indoor Residual Spraying.
*In all ages
Amongst children only
‡Estimated from the data provided
 
Table 4. Overview of studies comparing MDA + vector control versus no intervention

Study IDDesignCountryYearEndemicityMDA groupControl group/ baseline

Drug (dose)IntervalNo. of roundsPopulation
targeted
CoverageCo-intervention

Moderate Endemicity

Kaneko 2000N-RCSVanuatu1991-9929%*CQ (600 mg) + SP (1500 mg/75 mg) + PQ (45 mg) weeks 1, 5, and 9;

CQ (300 mg) + PQ (45 mg) weeks 2, 3, 4, 6, 7, and 8
1 week971879-92%ITN + larvivorous fishlow baseline coverage of ITNs

Ricosse 1959BASBurkina Faso1958-5915%Pyr (50 mg)2 weeks8500082-91%IRSNone

De Zulueta 1961BASUganda1959-6034%CQ (600 mg) + Pyr (50 mg)3 months430,38480%IRSNone

De Zulueta 1964BASUganda196023%CQ (600 mg) + Pyr (50 mg)5 months216,00050%IRSNone

Paik 1974aBASSolomon Islands197228%*CQ (600 mg) +Pyr (50 mg)1 month4ND90%IRSNone

High Endemicity

Molineaux 1980N-RCSNigeria1970-7546%*SP (500 mg/25 mg)10 weeks9‡14,12985%IRSNone

SP (500 mg/25 mg)2-10 weeks23‡181085%IRSNone

Escudie 1962N-RCSBurkina Faso1960-6156.1%CQ (600 mg) /AQ (600 mg) + PQ (15 mg)1 month8540081-92%IRSNone

CQ (600 mg)/AQ (600 mg) + PQ (15 mg)2 weeks15349082-94%IRSNone

Schneider 1961N-RCSBurkina Faso1960-6159%AQ (600 mg) + PQ (15 mg)2 weeks83525NDIRSNone

Metselaar 1961BASNew Guinea1958-5946%*CQ (450 mg) +Pyr (50 mg)1 week6250090%IRSNone

Hii 1987BASMalaysia1984-8546%SP (1500 mg / 75 mg) + PQ (30 mg)-175487%ITNNone

 N-RCS = Non-randomized controlled study; BAS = Uncontrolled before-and-after study; AQ = Amodiaquine; Pyr = Pyrimethamine; CQ = Chloroquine; SP = Sulfadoxine (or sulfalene)-Pyrimethamine; PQ = Primaquine; ND = Not described; IRS = Indoor Residual Spraying; ITN = Insecticide Treated Net.
*In all ages
Amongst children only
‡Estimated from the data provided
 
Table 5. Overview of studies assessing parasitaemia incidence only

Study IDDesignCountryYearBaseline IncidenceMDA groupBaseline

Drug (dose)IntervalNo. of roundsPopulation
targeted
CoverageCo-intervention

Garfield 1983BASNicaragua1981-820.4/1000CQ (500 mg/day for 3 days) + PQ (15 mg/day for 3 days)-12,300,00070-80%Larval controlNone

Simeons 1938BASIndia1935156/1000Ate (300 mg) + Plas (60 mg)-15650100%Larval controlNone

Gabaldon 1959BASVenezuela1956-570.4/1000Pyr (50 mg)1 week24111,995NDIRSIRS

Kondrashin 1985BASIndia19814/1000CQ (600 mg) + PQ (45 mg)6 months251,32585%IRSIRS

Paik 1974bBASSolomon Islands1972-7315/1000CQ (300 mg/day for 5 days) + PQ (15 mg/day for 5 days)3 months3120090%None-

Cáceres Garcia 2008BASVenezuela2002-0722/1000CQ (25 mg/kg over 3 days) +PQ (3.5 mg/kg over 7 days)-122,94177%None-

 BAS = Uncontrolled before-and-after study; PQ = Primaquine; CQ = Chloroquine; Pyr = Pyrimethamine; Plas = Plasmochin; Ate = Atebrin; ND = Not described; IRS = Indoor Residual Spraying.
Amongst children only
 
Table 6. Summary of findings table: Mass drug administration in areas of low endemicity (≤5%)

Mass drug administration in areas of low endemicity

Patient or population: People living in malaria endemic areas
Settings: Areas with low (≤5%) endemicity
Intervention: Mass drug administration (any regimen)
Comparison: Placebo or no intervention (or baseline data in before-and-after studies)

Timepoint
post MDA
OutcomesStudy designIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of
studies
Quality of the evidence
(GRADE)

Assumed riskCorresponding risk

ControlMDA

<1 monthParasite
prevalence
Before-and-after50 per 1000114 per 1000
(7 to 25)
RR 0.27
(0.14 to 0.50)
1 study⊕⊝⊝⊝
very low2,3,4,5

Parasite
incidence
----0 studies-

Gametocyte
prevalence
----0 studies2-

12 monthsParasite
prevalence
Before-and-after50 per 100011 per 1000
(0 to 6)
RR 0.02
(0 to 0.12)
1 study⊕⊝⊝⊝
very low 2,3,4,5

Parasite
incidence
----0 studies-

Gametocyte
prevalence
----0 studies2-

The assumed risk has been set at 5%. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk Ratio.

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 For illustrative purposes the control group prevalence has been set at 5%.
2 Only one cluster-randomized trial from Tanzania has evaluated MDA in a setting of low endemicity and this study recorded no episodes of parasitaemia or gametocytaemia at baseline or throughout six months follow-up in either the control or intervention groups.
3 Downgrade by 1 for serious risk of bias: This study is uncontrolled, and so at very high risk of confounding.
4 Downgraded by 1 for serious indirectness: This singe study from Taiwan reported the effects of MDA administered as a single dose of chloroquine (12 mg/kg). Further trials are needed from a variety of settings to have confidence in the results.
5 Compared to baseline data a large reduction in parasite prevalence was seen at 1 month and 12 months post-MDA.
 
Table 7. Summary of findings table: Mass drug administration in areas of moderate endemicity (6 to 39%)

Mass drug administration in areas of moderate endemicity

Patient or population: People living in malaria endemic areas
Settings: Areas with moderate malaria endemicity (6-39%)
Intervention: Mass drug administration (any regimen)
Comparison: No intervention (or baseline data in before-and-after studies)

Timepoint post MDAOutcomesStudy designIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of studiesQuality of the evidence
(GRADE)

Assumed riskCorresponding risk

Control MDA

<1 monthParasitaemia
prevalence
Non-randomized250 per 10005 per 1000
(3 to 15)
RR 0.03
(0.01 to 0.08)
3 studies⊕⊕⊕⊝
moderate1,2,3,4

Before-and-after250 per 100073 per 1000
(43 to 120)
RR 0.29
(0.17 to 0.48)
3 studies⊕⊕⊝⊝
low5,3,6

Parasitaemia
incidence
---0 studies-

Gametocytaemia
prevalence
Non-randomized100 per 100028 per 1000
(10 to 82)
RR 0.28
(0.1 to 0.82)
1 study⊕⊝⊝⊝
very low1,7

Before-and-after100 per 100047 per 1000
(25 to 87)
RR 0.47
(0.25 to 0.87)
3 studies⊕⊕⊝⊝
low5,6,8

4-6 monthsParasitaemia
prevalence
Non-randomized250 per 100070 per 1000
(53 to 95)
RR 0.18
(0.10 to 0.33)
2 studies⊕⊕⊝⊝
low1,3,9

Before-and-after250 per 1000438 per 1000
(103 to 1000)
RR 1.75
(0.41 to 7.41)
2 studies⊕⊝⊝⊝
very low5,10,11

Parasitaemia
incidence
----0 studies-

Gametocytaemia
prevalence
Non-randomized100 per 100052 per 1000
(24 to 111)
RR 0.52
(0.24 to 1.11)
1 study⊕⊝⊝⊝
very low12

Before-and-after100 per 100035 per 1000
(12 to 101)
RR 0.35
(0.12 to 1.01)
1 study⊕⊝⊝⊝
very low12

The assumed risk for parasitaemia prevalence has been set at 25%. Gametocytaemia prevalence was generally lower in the included studies and the assumed risk has therefore been set at 10%.

The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk Ratio.

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 No serious risk of bias: Although there were some differences in prevalence at baseline, these were much smaller in size than the large effects seen post-intervention.
2 No serious indirectness: These three studies were conducted in Kenya in 1953 and 1954 (pyrimethamine administered every six months for three rounds), and in India in 1953 (amodiaquine administered every two weeks for five rounds). A fourth study from Nigeria in 1973 reported a similar reduction in prevalence during an ongoing MDA program. Although these studies are old, similar effects might be expected today with effective anti-malarials.
3 No serious inconsistency: Consistent and large reductions were seen in these studies.
4 Upgraded by 1 for large effect size: Very large effects were seen consistently across both controlled and uncontrolled studies.
5 No serious risk of bias: These studies are uncontrolled, and so are at very high risk of confounding. However, as the GRADE approach automatically downgrades non-randomized controlled studies by two levels for risk of bias we did not further downgrade.
6 No serious indirectness: These three studies were conducted between 1953 and 1961, and administered MDA as: Pyrimethamine once only (Morocco), chloroquine plus pyrimethamine every month for five rounds (Nigeria) and chloroquine every four weeks for 11 rounds (Papua New Guinea). Although these studies are old, similar effects might be expected today with effective anti-malarials.
7 Downgraded by 1 for serious indirectness: This single trial in Kenya gave pyrimethamine every six months for three rounds. Different regimens may have different effects and primaquine, a drug with gametocytocidal properties, was not given. One further trial from Nigeria in the 1960s, which only reported on prevalence during an ongoing MDA programme, also administered MDA without primaquine.
8 No serious inconsistency: Gametocyte prevalence was lower post-intervention in all four trials, however there was variation in the size of this effect.
9 No serious indirectness: These two studies are both from Kenya in the 1950s, and both administer MDA as pyrimethamine alone. One study continued follow-up for > 6 months when an effect was still present.
10 No serious indirectness: These two studies were conducted between 1959 and 1961, and administered MDA as: chloroquine plus pyrimethamine every four months for two rounds (Cameroon), chloroquine plus pyrimethamine every month for five rounds (Nigeria).
11 Downgraded by 1 for serious inconsistency: At this time point results were mixed. One study found a higher prevalence at this time point and one found no difference.
12 Downgraded by 1 for serious indirectness: This single trial found no substantial difference between groups at 4-6 months. Modern trials with different regimens may have different effects. This study did not administer primaquine as part of MDA.
 
Table 8. Summary of findings table: Mass drug administration in areas of high endemicity (≥40%)

Mass drug administration in areas of high endemicity

Patient or population: People living in malaria endemic areas
Settings: Areas with high malaria endemicity (≥ 40%)
Intervention: Mass drug administration (any regimen)
Comparison: No intervention (or baseline data in before-and-after studies)

Timepoint post MDAOutcomesStudy designIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of studiesQuality of the evidence
(GRADE)

Assumed riskCorresponding risk

ControlMDA

< 1 monthParasitaemia
prevalence
Cluster-randomized500 per 1000410 per 1000
(335 to 505)
RR 0.82
(0.67 to 1.01)
1 study⊕⊕⊝⊝
low1,2,3

Non-randomized 500 per 100085 per 1000
(50 to 140)
RR 0.17
(0.10 to 0.28)
3 studies⊕⊕⊕⊝
moderate4,5,6,7

Before-and-after500 per 1000185 per 1000
(140 to 245)
RR 0.37
(0.28 to 0.49)
4 studies⊕⊕⊝⊝
low8,9,10

Parasitaemia
incidence
Cluster-randomized60 per 100025 per 1000
(14 to 44)
RR 0.41
(0.23 to 0.73)
1 study⊕⊕⊕⊝
moderate1,2,11

Gametocytaemia
prevalence
Non-randomized 100 per 100016 per 1000
(8 to 30)
RR 0.16
(0.08 to 0.30)
3 studies⊕⊕⊕⊝
moderate4,5,6,7

Before-and-after100 per 100038 per 1000
(13 to 108)
RR 0.38
(0.13 to 1.08)
3 studies⊕⊕⊝⊝
low8,12

4-6 monthsParasitaemia
prevalence
Cluster-randomized500 per 1000580 per 1000
(465 to 720)
RR 1.16
(0.93 to 1.44)
1 study⊕⊕⊕⊝
moderate1,2,13

Non-randomized ---0 studies-

Before-and-after500 per 1000205 per 1000
(120 to 360)
RR 0.41
(0.24 to 0.72)
3 studies⊕⊕⊝⊝
low8,14

Parasitaemia
incidence
Cluster-randomized60 per 100067 per 1000
(52 to 85)
RR 1.11
(0.87 to 1.41)
1 study⊕⊕⊕⊝
moderate1,2,13

Gametocytaemia
prevalence
Cluster-randomized100 per 1000107 per 1000
(62 to 185)
RR 1.07
(0.62 to 1.85)
1 study⊕⊕⊝⊝
low1,2,3

Non-randomized ---0 studies-

Before-and-after100 per 100035 per 1000
(10 to 128)
RR 0.35
(0.10 to 1.28)
2 studies⊕⊝⊝⊝
very low8,15

The assumed risk for parasitaemia prevalence has been set at 50%. Gametocytaemia prevalence was generally lower in the included studies and the assumed risk has therefore been set at 10%. The assumed risk for parasitaemia incidence is taken from the control group of the single trial.

The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
CI: Confidence interval; RR: Risk ratio.

GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

 1 No serious risk of bias: This cluster-randomized trial was at low risk of bias.
2 Downgraded by 1 for serious indirectness: This single study from the Gambia in 1999 administered MDA as AS+SP. The findings may not be easily generalized to other settings, or to alternative MDA regimens. The first time point measured post-MDA was 1-3 months.
3 Downgraded by 1 for serious imprecision: The result was not statistically significant but the 95% CI is wide and includes important effects.
4 No serious risk of bias: Although there was some evidence of baseline imbalance between the intervention and control areas, these were generally of smaller magnitude than the effects seen.
5 No serious indirectness: The data presented here were measured during ongoing multiple-round MDA programmes, not at one month post-intervention. The studies were conducted in Burkina Faso in 1961 (CQ or AQ plus PQ every two to four weeks), and Nigeria in 1975 (SP given every two weeks or every 10 weeks). Although these studies are old, similar effects might be expected today with effective anti-malarials.
6 No serious inconsistency: The observed effects were consistently large in all three trials.
7 Upgraded by 1 for the large effect size: Large effects seen in all trials.
8 No serious risk of bias: These studies are uncontrolled, and so are at very high risk of confounding. However, as the GRADE approach automatically downgrades non-randomized controlled studies by two levels for risk of bias we did not further downgrade.
9 No serious indirectness: These four studies were conducted in Palestine in 1930 (plasmoquine plus quinine every three weeks for three rounds), Burkina Faso in 1959 (pyrimethamine every two weeks), in Malaysia in 1985 (SP + PQ once only), and Cambodia in 2006 (AS + piperaquine once only plus PQ every 10 days).
10 No serious inconsistency: Three studies observed large effects, while one small study found no effect.
11 No serious imprecision: The result is statistically significant.
12 No serious indirectness: Two large studies found large effects in Burkina Faso in the 1950s (pyrimethamine every 2 weeks for 8 rounds), and Palestine in the 1930s (plasmoquine plus quinine every three weeks for three rounds). One small study from Malaysia in the 1980s found no effect.
13 No serious imprecision: The 95% CI excludes clinically important reductions at this time point.
14 No serious inconsistency: The two large studies from Palestine and Cambodia still demonstrated a large reduction at 4-6 months while the small study from Malaysia found no difference
15 Downgraded by 1 for serious indirectness: Benefits beyond three months have only been demonstrated in this single study from Cambodia. MDA was administered as artesunate plus piperaquine once only followed by primaquine every 10 days for six months.