Date: Friday, May 3, 1996                              Name: Ann Good Example

Journal Article Summary  

Authors: Thomas A. Mellman, Daniella David, Renee Kulick-Bell, and Joanne Hebding

Title: Sleep disturbance and its relationship to psychiatric morbidity after hurricane Andrew

Journal: American Journal of Psychiatry

Year: November 1995 Volume: 152 Pages: 1659-1663

The study was conducted in order to find out how a stress, such as Hurricane Andrew which

leads to psychiatric morbidity, can change sleep patterns.

The independent variables are Hurricane Andrew and the different psychiatric illness that

resulted because of the hurricane. The dependent variable is the amount/different sleep disturbances

that occur as a result of the stress of the hurricane. The experimenters were trying to answer the

question, will the stress of Hurricane Andrew cause people to suffer from sleep disturbances;

and if they did, did some who suffer from a psychiatric illness differ from others without disorders?

The subjects included 54 adults, 12 men and 42 women with an average age of 40.4 years,

who were living in areas that were directly affected by Hurricane Andrew. The control group

included 9 adults, 5 men and 4 women with an average age of 40.2, who were associated

with the affected areas.

There could be many ethical issues involved in this study due to how many lives the hurricane touched,

but the experimenters took time to have detailed interviews with each subject. If there was something

extremely wrong with a subject, they could have been referred for psychiatric help. With the information

collected, they can begin to treat many sleep problems. Subjects also signed an informed consent form.

There was no mention of staff except for those who interviewed the subjects and those who were trained

to interpret the data from the sleep laboratory. In the sleep laboratory, the data was recorded on cardiac

and respiratory monitors, EKG, EEG, and EMG. The Pittsburgh Sleep Quality Index was used to

assess sleep disturbance. In order to determine if someone was suffering from Post-Traumatic

Stress Disorder, they used the DSM-III-R symptom criteria. The interview structure was the Structured

Clinical Interview for DSM-III-R—Non-Patient Version.

The study was conducted in South Dade County, Florida during the clean-up of Hurricane Andrew.

The study of the sleep disturbances of the subjects was conducted in a sleep laboratory where the

subjects spent two days.

The first independent variable was Hurricane Andrew that destroyed areas of South Dade County,

Florida in August of 1992. Buildings were damaged due to 160 miles per hour winds and electricity

and water supplies were disrupted. The terror of the storm also slowed rescue and recovery efforts.

Once the storm was over, the stress began. The stress level rose as those affected tried to rebuild their

lives.  The second independent variable was the psychiatric illness suffered for those who witnessed the

trauma of Hurricane Andrew. The most common disorder was post-traumatic stress disorder. People also

suffered from depression and anxiety disorders.

First, the subjects were given an assessment determining what kind of psychiatric disorder they

suffered from. In order to ensure the reliability and validity of the results, they used the Structured

Clinical Interview for DSM-III-R—Non-Patient Version.  When they measured the patterns of

sleep disturbance before the assessment, they used the Pittsburgh Sleep Quality Index and then

added a modified version to find out about sleep patterns before the Hurricane. When the subjects

were put in the sleep laboratory, monitors and computers measured the sleep activity, this data

was recorded by trained technicians.

There was a large experimental group and that group was divided into those who suffered

from psychiatric morbidity and those who did not. The control group was much smaller.

The size of the control group is surprising due to the large experimental group.

Those with active morbidity suffered from bad dreams, global severity, sleep disturbances,

and awakenings. Those who suffered from active morbidity and those who did not have

above average scores that signified sleep disturbances, but those with psychiatric illness

scored significantly higher. These scores were based on a global severity. The average

was 5, but the people with active morbidity scored 9.5 while those without active morbidity

scored 5. When ten hurricane victims, or "experimental" group, were compared to the control

group, victims entered more often into state 1 sleep and had more arousal per hour.

Sketch the graphic representation of the results. (Just indicate trends in data.

A precise reproduction is not necessary.).

Discussion:

The results supported their hypothesis that that Hurricane Andrew brought about psychiatric

morbidity that led to sleep disturbances. They also found that "bad dreams" and post-traumatic

stress disorder are related.  This finding supported past experiments, although this was not a true

experiment and the experimental independent variable was not manipulated by an experimenter.

What is the major contribution of the study to the literature?

Since this article came from the American Journal of Psychiatry, it will be useful to all psychiatrists who

read this journal. Maybe with this information they can diagnose a patient’s sleeping disorder if they had

recently witnessed a trauma. It may also prompt other experiments dealing with this subject.

I thought this study was conducted very well. Although I had to receive help understanding the

terminology of some of the research terms, the article was overall easy to understand. There was

not much detail provided in the article regarding the study, but it seemed they did much research

even though they did not print the information.

  RETURN TO THE SYLLABUS