The title of the research is “The Association between Beta-Agonist use and Death from Asthma.” This study is based on a type of study called meta-analysis within various case studies. Meta-analysis involves the analysis of previously conducted studies, preferably those that have not been published to avoid publication bias. The authors of the study are Mullen Marylou, Mullen Brian, and Carey Michael. The major objective of the study is to establish, according to previous studies, the relationship between beta-agonist uses and death from asthma.
Beta-agonists are drugs used by most asthmatics to relieve the allergic reaction that causes the condition. Commercially, the drug is administered in form of bronchodilators and other methods to reduce the violent allergic reaction. Previous studies have shown that patients react differently to beta-agonist concerning age and mode of administration (metered-dose inhaler, nebulizer, and oral). Just the realization that there was a possibility that these drugs that have been long used for asthma can increase chances of death, is reason enough for study. However, the existence of many variables implies that several studies have to be conducted to validate this postulation beyond a reasonable doubt. The analysis provides the example of the 1992 research publication that was refuted as a mere scare by some scientists and authorities. Therefore, the study sought to analytical view the studies already conducted to at least shed more light.
The articles that were included in the analysis had to meet the following criteria:
All articles were from MEDLINE online library and the search was conducted using beta-agonist and asthma as the keywords.
Ancestry and descendency searching approaches were used.
All the studies that had been conducted before April 1992 were considered for the analysis.
Only the studies that recorded the exact number of cases and controls that received or did not receive beta-agonist were chosen.
These criteria led to the use of a total of six-case control studies made of fifteen separate tests addressing the relationship between beta-agonists and death from asthma, data involving 364 cases and 1388 controls were included. All studies that did not meet these requirements were excluded.
Analysis of Results
There was the general realization that most postulation dwelled on the fact that the relationship between death from asthma and the use of beta-agonists was restricted to the use of the nebulizer mode of administration. Therefore, there was a need to consider the mode of administration as a factor during the analysis of the studies. The other aspect of consideration was age. As established from various studies, children who learn to manage their asthma from a very young age may choose to review the methods used as compared to adults managing asthma. The studies also realized that in young children, there was always a high possibility of being assisted by parents. Therefore, in analyzing these studies these aspects had to be taken into consideration. Therefore, the research sought to gather all pragmatic information regarding the relationship between death from asthma and the use of beta-agonists. The information gathered will then be used to establish the extent to which the variables identified; age and mode of administration affected the relationship between death from asthma and administration of beta-agonists.
The study was based on a meta-analysis which refers to the ‘statistical integration of the results of previous studies’ (Mullen et al 1843). The procedural results of each study included are extensively analyzed concerning the hypothesis. In this particular study, x2 tests were obtained from each of the studies included that investigated the relationship between beta-agonists and death from asthma in terms of the candidates that used beta-agonists and the ones that did not use. These tests are then converted to the conventional statistical analysis for significance such as z and one-tailed p-test. The tests were also converted to effect sizes. Once these tests have been converted to more conventional measurements they can then be combined and used for predictive analyses. This method has been used extensively and found to be wanting. The method has been used by meta-analysis research conducted by Rosenthal and Rubin. The online MEDLINE computer search ensured that the study included most of the relevant literature addressing the relationship between the use of beta-agonists and death from asthma. x2 were obtained from, “four cell frequencies in the 2(case vs. controls) * 2(used vs. did not use beta-agonist), and this design allowed for direct derivation” (Mullen et al 1843). The result was that more than 200, both published and unpublished studies were used.
After organizing the chosen studies by their sample sizes, significance levels and effect sizes were conducted. The realization was that there was a significant relationship between the use of beta-agonists and death from asthma (z=3.996, p=0.000032). Although the realization was that there was a significant relationship, the analysis concluded that it was not strong enough. (mean r=0.55). The aspect of composite ratio was found to be 1.28. The use of a fail-safe number that ensured that no additional studies would be required to reduce the total probability to 0.5 was employed. The fail-safe number was found to be 36.18 which meant that 36 more studies had to be included to reduce the combined probability of p=0.000032 to 0.5. Analysis used by the Rosenthal approach indicated that the findings would significantly vary. The result was that a relationship was only obtained in instances when a nebulizer was used as the mode of administration, but the observation was that the relationship was rather weak. Adults were also found to exhibit a higher relationship between the use of beta-agonist and death from asthma than adolescents.
The conclusion arrived at was that in instances where a relationship was established, it still not right to make conclusions that the use of beta-agonists led to death from asthma. This is because the deaths could have been caused by some other inherent factors.
The study was conducted using most of the resources and methodologies applicable in this case and to this extent, the findings can be validated. However, there still looms the question of the cause of the relationship however weak.
The study allowed for the allowing for the findings concerning several studies in an area where conventional research would have been difficult. This is because most of the asthma patients used in the experiment might have already died.
There is no clear criterion for establishing the quality of the findings of several studies used. Furthermore, most studies have different objectives, and this may tend to impact the results when an analysis is conducted collectively.
Mullen et al. The Association between Beta-Agonist use and Death from Asthma: A Meta-Analytical Integration of Case Control Studies. JAMA, 1993.