1 / 24

Descriptive Study

Descriptive Study. Aya Goto. Key concepts. Nicely presented quantitative data tells a lot WITHOUT p-values. Graphs and tables. Grades of evidence. This is where research starts!.

basil-brennan
Download Presentation

Descriptive Study

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Descriptive Study Aya Goto

  2. Key concepts • Nicely presented quantitative data tells a lot WITHOUT p-values. • Graphs and tables

  3. Grades of evidence This is where research starts! Concato J, Shah N, Horwitz RI.Randomized, controlled trials, observational studies, and the hierarchy of research designs. N Engl J Med. 2000 Jun 22;342(25):1887-92.

  4. Descriptive and Analytical study • Descriptive study: Describe distribution of each factor individually. • Analytical study: Analyze associations among factors. In any study design, you must describe distribution of all factors you investigate. This is the very basic and most important of all study designs.

  5. Abortion rate(per 1000 women) 40 30 <20 20-24 25-29 20 30-34 35-39 10 40-44 0 1980 1985 1990 1995 Abortion trends in Japan, 1975-95. Goto A, Fujiyama-Koriyama C, et al. Stud Fam Plann. 2000; 31(4): 301-8.

  6. Abortion ratio (per 1000 live births) 10000 <20 20-24 25-29 1000 30-34 35-39 40-44 100 1980 1985 1990 1995

  7. Prevalence of and factors associated with reproductive tract infections among pregnant women in ten communes in Nghe An Province, Vietnam. Goto A, Nguyen QV, et al. J Epidemiol. 2005; 15(5): 163-72.

  8. Is protein causing colon cancer? Ecologic fallacy

  9. Age group Total <20 20-24 25-29 30-34 35-39 40-44 45-49 Abortion rate(per 1000 women) Japan 11.6 12.2 20.1 15.1 14.2 13.2 6.2 0.5 Fukushima 20.7 18.4 30.8 23.8 22.9 19.9 10.2 0.7 Abortion ratio (per 1000 live births) Japan 285 2249 512 154 156 420 1624 5775 Fukushima 390 2627 462 203 237 607 2324 13500 Table and figure format Table Table 1. Age-specific abortion rates and ratios in Japan and Fukushima, 2000

  10. Abortion rate (1000 women) 16.0 14.0 12.0 10.0 8.0 6.0 4.0 2.0 0.0 1995 1996 1997 1998 1999 2000 2001 Bar chart Figure 1. Trends in abortion rate, Japan, 1995-2001

  11. Abortion rate（1000 women） 25 20 15 Fukushima Japan 10 5 0 1980-84 1985-89 1990-94 1995-99 Line graph Figure 2. Trends in abortion rate, Japan and Fukushima, 1980-1999

  12. Histogram Figure 3. Age-specific proportion of abortion cases, Japan, 1999 <20 20-24 25-29 30-34 35-39 40-44

  13. USAGE • Table Precise data • Bar chart Trend • Line graph Trend • (Comparing several groups) • Histogram Distribution

  14. Summary measures Table 1. Basic characteristics of pregnant women * Electricity, bicycle, motorbike, running water, television, and telephone were the assets measured.

  15. Arithmetic rules can NOT be applied. •  N (%) • Categorical data • Blood type: 1= O, 2=A, 3=B, and 4=AB • Injury: 1=fatal, 2=severe, 3=moderate, 4=minor • Arithmetic rules can be applied. •  Summary measures • Continuous data • Hb level • Number of births that woman has given.

  16. Summary measures of continuous data • Mean=average • Standard Deviation(SD) • Mean±2SDs = a range in which “most” of your subjects fit. (About 95% lie within 2 SDs) • Median=50th percentile • Range=Min/Max • Mode=most frequent value(s)

  17. Age: 6, 6, 7, 8, and 25 • Mean = • Median = • Mode =

  18. Selection of Measures 1. Sample size: Large (>30) --> Mean (SD) Small --> Median (Min,Max) 2. Shape Median/Mode Mean Bell-shape  Mean (= Median = Mode) More than one peaks  Mode Skewed to the right (left)  Median

  19. Clinical descriptive epidemiologyModel case from a private clinic in JP

  20. Shironishi Clinic • Dr. Ogawa, Yutaka • General practitioner in a remote area

  21. Daily practice notes

  22. Referral notes

  23. Kosei No Sihyo 2000: 47(8): 23-28. “A systematic referral system should be established.”  A referral card was introduced in the region.  The paper was awarded.

  24. Simple clinical-oriented descriptive study Better practice

More Related