Pneumonia remains one of the most important causes of childhood mortality in developing countries. Treatment for five days with oral amoxicillin or trimethoprim-sulfamethoxazole is recommended for nonsevere cases. Many children clinically improve after the first few days of treatment, and parents stop antibiotic therapy prematurely. Qazi and colleagues note that children who discontinue antibiotic therapy early do not appear to have worse outcomes or more relapses than those who complete the full course. They conducted a randomized controlled trial comparing three to five days of amoxicillin treatment for acute pneumonia in children living in five Pakistani cities.

The authors recruited 2,000 children two to 59 months of age who met World Health Organization criteria for nonsevere pneumonia. The children had not recently used antibiotics and had no evidence of serious underlying disease. All children were treated with oral amoxicillin in a dosage of 15 mg per kg every eight hours for three days. For the remaining two days, one half of the children were given amoxicillin while the rest received an identical placebo. The children were assessed by a physician at three, five, and 14 days after enrollment. Respiratory rate, wheezing, and changes on chest radiography were the principal measures of clinical progress. Children who did not respond to amoxicillin therapy were treated with chloramphenicol for at least 48 hours, followed by injectable penicillin or oral cloxacillin for those who did not improve.

The failure rate in the three-day treatment group was 21 percent, compared with 20 percent in patients treated for five days. Overall, 243 children failed to respond after three days, and 113 of these responded to oral choramphenicol. Analysis of clinical failures showed that young age (11 months or younger), vomiting, at least three days of illness, and nonadherence to medication regimen were important factors. The rate of side effects was low and did not differ significantly between the two groups.

The authors conclude that three-day treatment is as effective as five-day treatment in non-severe childhood pneumonia. Very few children were lost to follow-up, and the rates of serious complications from the disease or the treatment were low. Shorter treatment with common antibiotics is effective without increased risk of relapse or worsening disease. The authors argue that short courses of treatment are likely to have higher rates of compliance and reduced risk of bacterial resistance, and offer substantial cost savings.