Page 24 - Fall 2008

tal noise levels have been rising consistently over the last 45
2
years. A compilation of published hospital noise levels indi-
cated that average day-time equivalent sound pressure levels (Leq) rose from 57 dB(A) in 1960 to 72 dB(A) in 2005, and night-time Leq levels rose from 42 to 60 dB(A). The Leq is defined to be the level of the continuous sound that would produce the same sound energy as the time-varying sound over a specified period. Many papers point out the lack of compliance with various standards and guidelines, such as those published by the World Health Organization (WHO),18 the U.S. Environmental Protection Agency (EPA),31 and the
32
very strict standards to meet.
Despite evidence of increasing awareness of the issue and
the potential for negative reactions of occupants, actual improvements have been sluggish at best in part because the pool of research remains limited in scope. The majority of articles that examine sound control strategies are mostly lim- ited to administrative noise controls such as closing doors
Overall hospital noise levels
An acoustical survey was conducted at a variety of wards in Johns Hopkins Hospital.2 Average equivalent sound levels (Leq), minimum sound pressure levels (LMin) and maximum sound pressure levels (LMax) as well as frequency distribution in octave bands were analyzed. A pediatric intensive care unit (PICU), children’s ward, oncology ward, and adult med- ical/surgical unit were included.
Results showed average Leq in the 50–60 dB(A) range. Corridors tended to be the noisiest areas, with nurses stations and occupied patient rooms the next noisiest. Empty patient rooms were significantly quieter. Of these types of spaces, only the empty patient rooms showed significant variations in noise levels as a function of time of day. On average, then, relatively constant sound levels were seen in areas that most impact patients, staff, and visitors.
The spectra were higher in the low frequencies (< 63 Hz), generally flat over the 63–2000 Hz octave bands, and followed by a gradual roll off above 2000 Hz. The low fre- quency energy was likely due to the heating, ventilating, and air-conditioning (HVAC) system. The flat sound spectrum region generally encompasses the speech band. This type of spectral distribution is not surprising given the high density of speech in hospitals. The high frequency energy was possi- bly due to alarms, mobile medical equipment, and high velocity airflow from the HVAC system.
During the course of this study, the opportunity arose to investigate overhead versus personal paging in the PICU. The PICU was originally dependent on overhead paging. Loudspeakers were active at least once every 5 minutes with each page typically lasting less than 30 seconds. An alterna- tive was implemented in the form of a personal, hands-free call system that broadcast only to the individual desired to be reached. As a result of the personal paging system, overhead pages in the PICU have been reduced to roughly one or two per hour.
Operating room noise
In another study, sound measurements were conducted in 38 operating rooms (ORs) to quantify the acoustical events during surgical procedures.38 Sound pressure levels were monitored before, during, and after a variety of operations on both adult and pediatric patients. The data were analyzed to determine the average Leq, peak sound pressure levels (LPeak), and frequency distribution in octave bands. Various classes of surgery were compared including cardiology, gastroin- testinal, neurosurgery, orthopedic, pediatric plastic, plastic, thoracic, and urology.
Example results are shown in Table 1. The Leq averaged between 55 and 70 dB(A) with significant sound peaks dur- ing surgical procedures. Orthopedic surgery had the highest Leq. The fraction of time that unweighted LPeakvalues exceed- ed 90, 95, 100, and 105 dB was also analyzed by category of surgery. This type of analysis provided insight into the “peak- iness” of sound in an OR during surgery, giving a much more detailed impression of the sound environment. For example, peak levels exceeded 100 dB over 40% of the time for neuro- surgery and orthopedic surgery. Peaks over 120 dB were not
The compi- lation showed that none of the previously published results complied with the WHO guidelines. It should be noted, how- ever, that these guidelines were intended to specify sound levels known to have no negative health impact—they are
American National Standards Institute (ANSI).
and asking staff to speak softly.
33,34
Such approaches have been
shown to have limited success in combating noise in indus-
try. Most of the previous work also has focused primarily on
overall levels of noise (i.e., Leq), but provide minimal exami-
nation of other detailed characteristics of sound such as the
frequency distribution, tonality, and time-variance. Previous
research has shown that these sound characteristics can
potentially impact our reaction to the soundscape by causing
annoyance, decreasing performance, eliciting physiological
35-37
standing the problematic nature of the hospital soundscape. However, additional information on the psychological and physiological reactions of occupants to detailed properties of the acoustical environment is needed to improve conditions. To advance the state of knowledge, a research collaboration involving engineering, architecture, and medicine has been formed over the last several years.
Recent research on the hospital soundscape
One important aspect to consider in hospital or health- care facility research is the uniqueness of the different spaces housed therein. Variations in the architecture, types of equip- ment, conditions of patients, presence of visitors, occupa- tional culture, and occupant activities can be found when comparing different types of wards. In our work, we are endeavoring to take a comprehensive look at different types of wards. Together with our collaborators, we have thus far examined spaces including operating rooms, hematological cancer units, burn acute care units, adult emergency depart- ments, and a variety of intensive care units (ICU). Hospitals that have been included are Johns Hopkins Hospital in Baltimore, MD, Emory University Hospital in Atlanta, GA, Parkland Hospital in Dallas, TX, and the Sahlgrenska University Hospital in Gothenburg, Sweden.
reactions, etc.
The existing literature provides a good basis for under-
Evaluating the Hospital Soundscape 23