Miller 2006

Methods	Trial design: Open‐label, randomized, controlled trial with three arms, followed by an open‐label study with one arm. Follow‐up: Hospitalized, medical history and examination on enrolment. Daily history of signs, symptoms and medication. Routine physical exam on "a periodic schedule". Blood smears twice daily until malaria cleared, then daily for a week, then weekly or as clinically warranted until day 28. Adverse event monitoring: All patients were evaluated daily during treatment for adverse events that were new in onset or aggravated in severity or frequency after administration of the study drugs. An adverse event was considered drug related if its relationship to treatment was rated definite or probably by a study clinician. Complete blood count, urinalysis, blood urea nitrogen, creatinine, glucose, ALT, gamma‐glutamyl transferase, urine pregnancy test (women only), and ECG at enrolment. Not clear when laboratory tests were repeated ("on a periodic schedule"). Adverse events reported for those treated.
Participants	Number enrolled:  61 Number available for analysis: 60 Inclusion criteria:  age ≥20 years; males and non‐pregnant females, visiting the OPD with fever; Plasmodium falciparum infection; written informed consent. Exclusion criteria: Known history of chronic illness; signs or symptoms of severe or complicated P. falciparum malaria (WHO 2000) or hyperparasitaemia (> 5% of red blood cells infected), or sustained hyperpyrexia (fever > 40 °C); significant liver dysfunction (ALT > 300IU/l); known pregnancy or positive urine test result for betha‐hCG; mixed malaria infection by Giemsa smear; drug therapy for P. falciparum in the previous 42 days; history of allergy to study medicines or concurrent use of drugs with known interactions with study drugs. Age range: 20 to 60 years, mean 36 (arm 1), 31 (arm 2), 35 (arm 3), 35 (arm 4) Proportion male: 44/60 (73.3%) Pregnant women: Excluded
Interventions	arm 1) Azithromycin 500 mg two times per day for three days plus quinine 10 mg/kg three times per day for three days. arm 2) Azithromycin 500 mg two times per day for five days plus quinine 10 mg/kg three times per day for five days. arm 3) Doxycycline 100 mg two times per day for seven days plus quinine 10 mg/kg three times per day for seven days. arm 4) Azithromycin 500 mg three times per day for three days plus quinine 10 mg/kg three times per day for three days: This arm was evaluated a rainy season after the other arms without comparator drug (so non‐randomized) Supervision of intake: Yes Other (symptomatic) treatment:  Acetaminophen and dimenhydrate provided as needed for symptoms by participant. Other medications or intravenous fluids as prescribed by a physician
Outcomes	Outcomes used in this review:  Total failure by day 28 (primary endpoint), PCR‐adjusted failure rate by day 28 Parasite and fever clearance time (secondary outcomes) Adverse events
Notes	Location:  Armed Forces Research Institute of Medical Sciences inpatient clinical trials centre, Kwai River Christian Hospital, Sanghlaburi district, Kanchanaburi Province, Thailand (western border with Myanmar). Date:  July 1998 to October 2001 Malaria endemicity:  Low seasonal transmission Drug resistance: Resistance in study area reported to chloroquine, sulphadoxine‐pyrimethamine and mefloquine Method of determining parasite density:  Giemsa stained thick and thin smears; count of asexual and sexual parasites per 200 WBCs; if less than 10 parasites, counted against 500 WBCs. Negative if no parasites in 200 high‐power fields. Smears read on site, and re‐read by off‐site blinded microscopist. Parasite density was the mean of non‐discrepant values. Discrepant results decided by third microscopist. Baseline data: Gender, age, P. falciparum parasite density, temperature, baseline laboratory data (white blood cell count, red blood cell counts, platelets, kidney and liver functions), reported by treatment arm for those analysed. Look similar.  Drug source:  Quinine sulphate: Zenith Goldline Pharmaceuticals, Miami, Fl, provided by Walter Reed Army Medical Center (Washington DC); Doxycycline hyclate: Qualitest Parmaceuticals, Huntsville, AL, provided by Walter Reed Army Medical Center (Washington DC); Azithromycin: produced and provided by Pfizer Inc, New York, NY. Source of funding: Pfizer Inc. and U.S. Army Medical Research and Materiel Command. Other notes:  Enrolment in high dose azithromycin arm in later (different) malaria season and without comparator arm (historical comparison), so might not have been randomized. Facility had screens on windows, so low risk on re‐infection inside the facility.
Risk of bias
Bias	Authors' judgement	Support for judgement
Adequate sequence generation?	Unclear risk	A block randomization table of 10 QAZ to 2 QD subjects was generated. To improve power of the study, enrolment to a well‐tolerated QAZ arm was increased with 10 (year 1). In the subsequent year's malaria season (year 2), an additional regimen with a higher daily azithromycin regimen was assessed without comparator. Thirty women were enrolled in year 1 in an azithromycin regimen, but there were 10 in the QD arm, whereas there should have been six. Volunteers withdrawing before trial completion were replaced in the randomization schedule (applicable to one participant in an azithromycin arm).
Allocation concealment?	Unclear risk	Not reported if and how allocation was concealed
Blinding? All outcomes	High risk	Microscopists were blinded to treatment regimen (reported by authors). Otherwise open label
Incomplete outcome data addressed? All outcomes	Low risk	Low loss to follow‐up: one participant elected to withdraw on day two and was replaced
Free of selective reporting?	High risk	One arm was not at the same time as the other three regimens (not randomized)
Free of other bias?	Unclear risk	Not clear how much study sponsor was involved in reporting of results and adverse events