But, it must be said, he failed in his hand-washing crusade. Sherwin Nuland, award-winning writer and professor of surgery at Yale University, details the many reasons behind Semmelweis's failure in the book, The Doctors' Plague: Germs, Childbed Fever, and the Strange Story of Ignác Semmelweis (2003). Aside from resistance among his colleagues at the Vienna General Hospital--which doctor, after all, given the body of knowledge available at the time, would be prepared to see himself as the cause of too many deaths?--Semmelweis was reluctant to systematically document his findings. He was plagued with a "greater sense of unworthiness" and was invariably combative and rude to colleagues who questioned him. He also became progressively mentally unstable, leading to a breakdown. He was then committed to an institution, where, although he was reported to have died of an infection, Nuland says he was assaulted to death by the asylum staff.

    Some years before his death Semmelweis wrote: "Because of my convictions, I must here confess that God only knows the number of patients who have gone to their graves prematurely by my fault."

From strange to standard

    Semmelweis's story may just be one of the "strange" stories in medical history, except for the fact that his discovery has become one of the enduring standards in medicine. Postpartum-sepsis-related deaths are no longer as widespread now as in the 19th century; still, infections in general are a continuing challenge in health care.

    It goes without saying that hand hygiene remains a key factor in preventing the spread of infections, particularly in the hospital setting. Nosocomial pneumonias and sepsis, for example, remain a common problem in most hospitals. This is especially critical for newborns.

    Dr. Jacinto Blas Mantaring, neonatologist and associate professor of clinical epidemiology at the University of the Philippines Manila, said that even today, infections seriously threaten the lives of newborns. "There was a time," he said, "whenever a newborn was diagnosed to have an infection, his chances of survival were already 50 percent." In addition, infection rates in the nurseries of many hospitals in the country continue to be high, with some rates going between 20 percent and 30 percent.

    "It is, theoretically, very easy to pass an infection from one baby to the other," observed Mantaring. One of the factors that could make the transmission of infections in the nursery is the "very poor" nurse-to-patient ratio in government hospitals--the ideal ratio is one nurse for every two or three patients. "In the nursery, we have one nurse to 10 patients," he said.

    The hand hygiene of those involved in handling the babies is another critical factor. While simple hand washing is established as an important way in lowering the risk of transmitting infections, how strictly the hospital staff sticks to this standard could be interpreted in several ways. Also, repeated hand washing could eat a lot of a doctor or nurse's time, given that he or she has to see 10 patients at a time.

    And since increasing rates of infection result in increasing rates of antibiotic usage, the risk for antimicrobial resistance also rises.

"A simple set of interventions"

    With the goal of decreasing the rates of infections and deaths among newborns, Mantaring, together with colleagues from the University of the Philippines National Institutes of Health, Philippine General Hospital, Jose Fabella Memorial Hospital, and Boston University School of Public Health, conducted a study determining the effectiveness of a "simple set of interventions" in the neonatal-intensive-care unit.

    The study was conducted in two sites--the NICU at the PGH and Fabella. The importance of hand hygiene was emphasized in the study. An alcohol-based hand wash (80-percent solution) was introduced. Explained Mantaring: "At best hand washing can remove maybe 80 percent of germs, but it has been shown that an alcohol-based hand rub can remove 90 to 95 percent."

    Also, a "feedback" system was set up to ensure hand-hygiene compliance, and this primarily involved frequent reminders. During observation--the time and the beds for observation were randomly selected--a research assistant listed down all encounters during that time and near those beds. The assistant noted who was involved with the encounter, what was performed, and whether hand-hygiene practices were being adhered to. Seminars also regularly conducted to emphasize the importance of washing one's hands, or using the alcohol-based hand rub made available at the bedside.

    The study had two phases: a preinter-vention phase (lasting five months at PGH and 10 months at Fabella) and an intervention phase (10 months at PGH and five months at Fabella). In both phases, bacterial-colonization rates (through samples taken from the perirectal, umbilical, and nasal areas) were taken, as well as bloodstream-infection rates and overall mortality.

Significant reduction in deaths

    There was an increase in compliance with hand-hygiene practices. While during the baseline phase consistent hand washing and use of the hand rub had only been between 10 and 20 percent, the intervention phase saw the rates rising to between 30 and 40 percent. Mantaring also pointed out that doctors and nurses tended to be "a little selective" of when or when not to be especially careful with their hand hygiene. "If they were just taking vital signs, [for example], then they were less conscious," he said. If the procedure were more sensitive, compliance was much higher. He also clarified that the 30 to 40 percent compliance rate could be an "underestimate," because the more serious tasks could bring a compliance rate of upwards of 90 percent.

    Meanwhile, colonization and mortality rates at baseline were high; some of the babies had been shown to be colonized as early as the first day of life, "which somehow speaks of either the hygiene of the mother or the hygiene of the obstetric-admitting section," said Mantaring.

    By the end of the study period, sepsis-related and all-cause mortality went down significantly. At the PGH, deaths were practically halved--from the 292.7 deaths per 1,000 admissions during baseline, the case-fatality rate dropped to 163.9 per 1,000 admissions on the 10th month of intervention. At Fabella, meanwhile, the case-fatality rate went down from 659.2 per 1,000 during baseline to 543.6 per 1,000 at the fifth month of intervention.

    Despite the significant reduction in deaths, the colonization and sepsis rates were practically unchanged. How, then, could these have brought about a reduction in deaths? Mantaring and colleagues believed that it could have been brought about by the decrease in two specific microbes--Alcaligenes faecalis and Pseudomonas aeruginosa. "It was more a qualitative change than a quantitative change," he said.

    The study, therefore, just confirms what Semmelweis posited in the 1840s--proper hand hygiene is the first best weapon against infections and deaths resulting from infections. Especially in settings where resources are limited, hand washing, supplemented with a readily available alcohol-based hand rub, is a simple, inexpensive, and effective intervention.