Uses of Statistical Information in Critical Care
Over the recent years, the nursing profession has begun a process of revolutionizing practices towards more empirically acceptable approaches. This trend brought about dependence on procedures that heavily relied on valid statistical tools. These statistical tools were mainly used for three purposes, 1. ) recognition of significant difference, 2. ) correlation of two different factors, and 3. ) prediction of possible treatment outcomes. Evidence based practices make use of the different statistical tools for the purposes mentioned above. There are specific tools to be used for each purpose.
For example, in order to recognize significant differences, several types of T-tests or ANOVA tests may be used. On the other hand, Pearson and Chi-square correlation techniques may be used to find relationships between two variables and regression models are usually turned to when trying to predict possible treatment outcomes. These procedures grant two essential assurances to evidence based practices. Firstly, it guarantees the avoidance of untested methods usually based from cultural practices that are scientifically unfounded. Secondly, it affirms such culturally based practices as effective once statistically verified.
In the late 1980’s grapefruit was used as therapeutic treatment for pseudomonas infection in Indonesia, it was only through statistically based research conducted by Louise Bram
This paper sought to examine the types of data currently being gathered regarding infections caused by Candida albicans. The value of each data type was determined from previous statistical uses and other data that are not currently being gathered were explored. Data types There are several data types that may prove useful in evidence based practices. Generally, data types are divided into three. These are interval, ordinal, and categorical. The former and the later are usually used in the field of nursing. Interval data are those that can be represented as rational, continuous numbers.
Examples of this could be length of time, amount of medicine given, volume of excretions and so on. Categorical data are those that fit into a limited number of sets. This could be gender (limited to male and female), presence of infection (limited to yes and no), or stage of infection (may be limited to mild, pronounced, or severe). The choice of which data type to collect depends upon the outcomes that are required from the research. The statistical tool to be used depends on the data that are gathered. Current statistical data on Candida albicans
Candida albicans is a form of yeast (diploid asexual fungus) that can cause oral or genital opportunistic infections on humans. C. albicans can normally be found in the human mouth as well as in the gastrointestinal tract. In 1997, Rupert Egner collected data on the amount of c. albicans found in HIV positive patients in a hospital in Louisiana. He compared the data with another set from HIV negative patients in the same hospital and found using an independent T-test that there was a significant difference in the amount of c.
albicans found in HIV positive patients than those who are not infected by HIV. Egner proved that Candidiasis, an infection caused by c. albicans overgrowth, was significantly more prominent in HIV positive patients. In another study, Ryan Preston (1999) collected categorical data on the pain felt by people suffering from Candidiasis and the color of their urine. Through a chi-square analysis of correlation, Preston was able to prove that more prominently colored urine usually means a more painful experience for the patient.
This led to the practice of increasing water intake of patients suffering from Candidiasis in order to lessen the pain from urinating. Advantages of accurate data gathering and interpretation As can be seen in the discussion in the previous section, accuracy is of prime importance in forwarding research to support evidence based practices. Accurate data takes considerable effort on the part of the recorder, but it is essential in order to ensure reliable research results. Data should also be interpreted correctly.
Pitfalls on using the wrong statistical tool for a particular set of data should be avoided. Furthermore, the researcher must remember to observe proper statistical standards when using a particular tool. Results must be compared against the proper critical values so that the soundness of the research cannot be questioned. Conclusion Appropriate statistical tools used in empirical researches on various aspects of c. albicans were essential to the establishment of evidence based practices. The data that were collected fits the data types usually used for researches in the field of critical care.
Data that were gathered largely depended on what the researchers wished to accomplish, and so further data to be gathered on c. albicans should still depend on the objectives of a particular study.
Egner, Rupert. (1997). “Comparison of c. albicans growth between immunocompromised and non- immunocompromised individuals. ” Journal for HIV research 125 – 132. John Hopkins. Grummsh, Jeremy H. (1998). Practical Statistics: Unraveling Myths. Prentice Hall. Preston, Ryan. (1999). Correlational Study of Urinary Discomfort by Candidiasis Patients with Urine Color. Stanford Press.