Relationship Between ANA Test Titers Autoimmune Disease - Research Paper Example

Relationship between ANA Test Titers autoimmune Disease Results Among all 44 patients, majority, [18 (40.9 were Qataris, followed by Indians [6 (13.6%)], then Egyptians [5 (11.4%)], then Sudanese [4 (9.1%)] were Sudanese, then Filipinos and Indonesian [2 (4.5%)], and then Algeria, Canada, Hungary, Iran, Palestine, Sri Lanka and Yemen [1 (2.3%)] each. There were 35 (79.5%) females and 9 (20.5%) males. Among these 44 patients 2 (4.5%) were below 10 years, 12 (27.3%) were between 10 years and 20 years, 6 (13.6%) were between 20 years and 30 years, 11 (25%) were between 30 years and 40 years, 7 (15.9%) were between 40 years and 50 years, and 6 (13.6%) were above 50 years. Among the patients ‘data we collected, 13 (29.5%) patients had 40 ANA test titers, 4 (9.1%) had 80 titers, 4 (9.1%) had 160 titers, 10 (22.7%) had 320 titers, 4 (9.1%) had 640 titers, 6 (13.6%) had 1280 titers, and 3 (6.8%) had 1320 titers. This is well illustrated by both table 1 below and graph 1. Table 1: Distribution of the 44 patients among the ANA tests titers ANA test SLE 40 80 160 320 640 1280 1320 Total 13 4 4 10 4 6 3 % 29.5% 9.1% 9.1% 22.7% 9.1% 13.6% 6.8% Figure 1: Distributionof the 44 patients among the ANA tests titers During the sample, analysis three groups of diagnosis of autoimmune disease were detected. Group 1 had 13 (29.5%) patients, group 2 had 19 (43.2%) patients, and group 3 had 12 (27.3%). Table 2 and figure 2 better illustrate this Table 2: distribution of all the 44 patients among the three groups of diagnosis Group Group 1 Group 2 Group 3 No. of patients 13 19 12 % 29.5% 43.2% 27.3% Figure 2: Distribution of all the 44 patients among all the three groups of diagnose Among the 13 patients who were found to fall in the category of 40 titers ANA test SLE, 2 were found to fall in group 1, 6 were found to fall in group 2, and 5 were found to fall in group 3. Among the 4 patients who were found to fall in the category of 80 titers ANA test SLE, 2 were found to fall in group 1, 1 one was found to fall in group 2, and another 1 was found to fall in group 3. Among the four in the category of 160 titers ANA test SLE, 2 were found to fall in group 1, 1 in group 2, and another 1 in group 3. Among the 10 in the category of 320 titers, ANA test SLE, 3 were found to fall in group 1, 4 in group 2, and 3 in group 3. Among the 4 in the category of 640 titers ANA test SLE, 1 was found to fall in group one while the other 3 were found to fall in group 2. Among the six in the category of 1280 titers ANA test SLE, 2 were found to fall in group 1, 2 in group 2, and 2 in group 3. Among the three in the category of 1320 titers ANA test SLE, 1 was found to fall in group 1, and the other 2 in group 2. Table 3, figure 3, and figure 4 had better illustrate this. Table 4: Distribution of the patients among the categories of the ANA tests SLE and the groups ANA test SLE 40 80 160 320 640 1280 1320 Total % Group 1 2 2 2 3 1 2 1 13 29.5% Group 2 6 1 1 4 3 2 2 19 43.2% Group 3 5 1 1 3 0 2 0 12 27.3% Total 13 4 4 10 4 6 3 44 100% % 29.5% 9.1% 9.1% 22.7% 9.1% 13.6% 6.8% 100% Figure 3: Distribution of the patients among the categories of the ANA tests SLE and the groups Figure 4: pie charts Discussion In this project, we test the null hypothesis that the ANA and diagnosis are associated. According to Hirschfield & Heathcote (2011), the ANA test is the main testing tool used for the diagnosis of autoimmune rheumatic conditions. This test makes use of antinuclear antibodies in the diagnosing of these conditions. “Antinuclear Antibodies (ANA) are heterogeneous group of auto antibodies directed against nucleic acids and nucleon proteins within the nucleus and cytoplasm” (Lee, 2009, p. 452). In this experiment were have analyzed the ANA test titers and compared them with the group of the diagnosis. The ANA titer measures the total amount of serum antibodies in a person and it should be strongly positive for us to confirm the diagnosis (Mutasim & Adams, 2000). As the data in table one show, different ANA test titers were recorded for the 44 patients in our sample. This can be used to show that ANA titers differed among patients regardless of the group of the diagnosis. However, there were equal number of patients with 80, 160, and 640 ANA test titers was the same. This can be used to show that there is some relationship between these ANA test titers. Hence, a statistical analysis is necessary to confirm this relationship. As the data in table 2 indicate, there is different number of patients in each of the diagnosis group. In this case, there is no group with equal number of patients and hence we can conclude that there is no any relationship between the groups. The data in table 3 shows that there is different number of patients who fall in the same category of ANA test titers and the same group of diagnosis. All the categories that can be derived from this table have different number of patients. This is more clearly revealed by figure 3 in the results. This makes us suspect that there might be no relationship between the ANA test titers and the diagnosis. However, we need to confirm this using statistical analysis on the data in order to conclude whether our null hypothesis is true of false. In our statistical analysis we did a chi-square test on our data, followed it with model fitting and then the likelihood ratio test. This is the most appropriate test when your data is group in two different ways (Albright et al. 2010). The purpose of the chi-square test was to check whether we should accept or reject the null hypothesis. The model fitting was to check how well the data fits in a statistical model to support the conclusion of the null hypothesis. In addition, the likelihood ratio test was to confirm the information derived from the model fitting. According to Rice (2007), likelihood ratio test is very much recommended in testing of hypotheses. In our statistical analysis we use a level of significance of .001, hence we would accept the null hypothesis when p> .001 and otherwise reject it. In our statistical analysis, we found our chi-square test statistic to be 6.835 with a p- value of 0.868. Clearly, this p-value is greater than our level of significance that is 0.001. We should thus accept the null hypothesis and conclude that ANA test is not associated with the diagnosis. However, this conclusion has to be supported by its proceeding analysis. Since, the chi-square statistic is the difference in two log-likelihoods between the final model and a reduced model, the null hypothesis is that all parameters of that model should be zero. From this model fitting we find out omitting the effects does not increase the degrees of freedom and thus the final model and the reduced model are equivalent. This is confirmed by the likelihood ratio test in which we obtain a chi-square statistic of 8.558 with a p-value of .740 which is greater than 0.001. According to a research undertaken in Qatar a female patient was diagnosed with abdominal pain and ascites, mesenteric lymphadentis and peritoneal panniculitis but initially but her ANA test initially turned up to be negative. This research also found out that ANA test turned positive in 97% of the SLE patients involved in the research but only 21% were diagnosed with the Behcet’s disease (CAGS, 2011). These research findings showed that there is no association between ANA test titer and the diagnosis of this disease. Another research undertaken in Egypt showed that only a complete profile of ANA is the one that might be used in the diagnosis of patients suspected to have autoimmune liver disorder (HPPD, 2009). Limitations In this research I encountered several problems one of them being the change of the laboratory location without my awareness. Another limitation was the time allocated for the project. We had been allocated only two month to do the project and since this project required collection of large samples of data, the two moths allocated was a challenge. Another limitation was that the medical records were hand written and it was not found online so I had to look for them from the hospitals. Moreover, I encountered another limitation of having some of the files where I could derive the data being not available due to the hospitals policies of storing files in another facility if the file was not used for a period of five months. Summary and Conclusion From my analysis, I can conclude that the ANA tests are not associated with the diagnosis. This can also be confirmed by the two researches that we have looked at one from Qatar and the other one from Egypt. However, according to Mutasim & Adams (2000), the diagnostic value of an ANA test is dependent of the clinical setting. Thus, we cannot be certain whether this conclusion can hold in all the clinical setting. In this experiment, we were only able to obtain the data of forty-four patients whose ANA titers were tested. This is not a large enough sample to avoid any kind of biasness in the data. As a result, the statistical data might have a certain degree of biasness and thus might be giving inadequate of misleading information. Therefore, I propose that another larger supple be undertaken to test this hypothesis and confirm what is obtained from this analysis. Reference Albright, S. C. et al. (2010). Data Analysis and Decision Making. Mason, OH: Cengage Learning. CAGS. (2011). Systematic Lupus Erythematosus. Retrieved from: http://www.cags.org.ae/pdf/152700.pdf. On 8 June 2012. Hirschfield, G. M., & Heathcote, E. J.(2011). Autoimmune Hepatitis: A Guide for Practicing Clinician. New York, NY: Springer. HPPD.(2009). Index Medicus for the WHO Eastern Mediterranean Region: With Abstracts. Retrieved from: http://applications.emro.who.int/dsaf/dsa997.pdf. On 8 June 2012. Lee, M. (2009). Basic Skills in Interpreting Laboratory Data.Berlin: Springer. Mutasim, D. F., & Adams, B. B. (2000). A Practical Guide for Serological Evaluation of Autoimmune Connective Tissue Diseases. Retrieved from: http://faculty.washington.edu/andchien/PDFs/autoantibodies.pdf. On 8 June 2012. Rice, J. A. (2007). Mathematical Statistics and Data Analysis. Mason, OH: Cengage Learning. Read More