UKURAN ASOSIASI DALAM EPIDEMIOLOGI Putri Handayani, M. KKK
Tipe ukuran yang digunakan dalam epidemiologi • Ukuran asosiasi – Merefleksikan kekuatan atau besar asosiasi antara suatu eksposur/faktor risiko dan kejadian suatu penyakit – Memasukkan suatu perbandingan frekuensi penyakit antara dua atau lebih kelompok dengan berbagai derajat eksposur – Beberapa ukuran assosiasi digunakan untuk mengestimasi efek
Basic Question in Analytic Epidemiology
• Are exposure and disease linked?
Exposure
Disease
Ukuran-ukuran asosiasi • Ukuran rasio (perbandingan relatif) – rasio dua frekuensi penyakit membandingkan kelompok terpajan dengan kelompok tidak terpajan
• Ukuran perbedaan efek (perbandingan absolut) – perbedaan antara ukuran frekuensi penyakit suatu kelompok terpajan dan kelompok yang tidak terpajan
Two-By-Two Table Disease
Yes
Exposure
No Total
Yes
No
Total
a
b
a+b
c
d
c+d
a+c
b+d
a+b+c+d
Hypothetical Two-By-Two Table Lung cancer
Yes
Smoking
Yes
No
Total
70
300
370
15
700
715
No Total
85
1,000
1,085
Relative Risk (RR) • Measures how likely the exposed group will develop a disease compared to the unexposed group.
RR =
incidence in the exposed incidence in the unexposed
= a/(a+b) c/(c+d)
Example: Hypothetical Study Lung cancer
Yes
Smoking
No
Total
Yes 70
300
370
No 15
700
715
1,000
1,085
Total 85
Relative Risk = 70/(70+300) = 9.0 15/(15+700) Which means… participants who smoked were 9 times more likely to develop lung cancer than those who did not smoke.
Interpreting Measures of Association RR of 1.0 indicates that the occurrence of disease in the exposed an unexposed groups are identical: –No association observed between exposed and unexposed groups.
Interpreting Measures of Association (Continued) • RR greater than 1.0 indicates a positive association, or an increased risk among the exposed. • RR less than 1.0 means that there is a decreased risk among the exposed group.
Ukuran-ukuran asosiasi • Ukuran rasio – Rasio risiko atau risiko relatif (RR) RR
Risiko pada kelompok terpajan Risiko pada kelompok tidak terpajan
– Rasio Insidens Kumulatif (RIK) RIK
Insidens kumulatif pada kelompok terpajan Insidens kumulatif pada kelompok tidak terpajan
Perhitungan RR untuk CI Outcome (+)
Outcome (-)
Total
E (exposed)
a
b
a+b
NE (unexposed)
c
d
c+d
Total
a+c
b+d
N (a + b + c + d)
CI pada populasi exposed (E) --- a/(a + b) RR= CI pada populasi unexposed (NE) --- c/(c + d)
THEN,
FIRST, SELECT
FOLLOW TO SEE WHETHER
Incidence rates of Disease
Disease Develops
Disease Does Not Develop
Totals
Exposed
a
b
a+b
a Incidence in = exposed a+b
Not Exposed
c
d
c+d
c Incidence in = c+d non-exposed
Relative Risk ( RR ) = incidence in exposed incidence in non-exposed =
a a+b c c+d
Example :
A Prospective Study of 3,000 Smokers and 5,000 Nonsmokers to Investigate Smoking and Coronary Heart Disease (CHD) THEN FOLLOW UP TO SEE HOW MANY
Healthy Smokers FIRST, SELECT
Healthy Non-Smokers
Develop CHD
Do Not Develop CHD
Total
Incidence per 1,000 per year
84
2,916
3,000
28.0
87
4,913
5,000
17.4
Contoh 5. Tabel 1. Kaitan antara merokok dan angka insidens stroke dalam suatu kohort. Kategori merokok
Tidak pernah merokok Mantan perokok Perokok
Total
Jumlah kasus stroke
Orang-tahun observasi (lebih dari 8 tahun)
Tingkat insidens stroke (per 100.000 orang tahun)
70
395.594
17,7
65 139
232.712 280.141
27,9 49,6
274
908.447
30,2
Sumber: diterjemahkan dari:Beaglehole et al. Basic Epidemiology. WHO. 1993. 18.
Postmenopausal Hormone Supplement and CHD
Ever use Past use Current Never use
CHD
Person-years
30 19 11 60
54,308.7 24,386.7 29,922.0 51,477.5
RR ever use vs never use = (30/54,308.7) / ( 60/51477.5) = 0.5 RR past use vs never use = (19/24386.7) / (60/51477.5)
= 0.7
RR current use vs never use = (11/29922.0) / (60/51477.5) = 0.3
Ukuran-ukuran asosiasi • Ukuran rasio – Rasio odds (Odds ratio = OR) • Nama lain: Odds relative; rasio kros-produk • rasio dua odds yang digunakan dalam studi kasuskontrol untuk mengestimasi rasio rate atau rasio risiko
Ukuran-ukuran asosiasi • Ukuran rasio – Rasio odds (Odds ratio = OR) • odds untuk satu kelompok dibagi dengan odds untuk kelompok yang lain • Mempunyai interpretasi yang sama seperti risiko relatif
Ukuran-ukuran asosiasi • Odds suatu kejadian – rasio probabilitas bahwa kejadian terjadi terhadap probabilitas kejadian tidak terjadi Odds suatu peristiwa
P 1–P
P 1 P
= Probabilitas suatu kejadian terjadi = Probabilitas suatu kejadian tidak terjadi
Figure 11-5 A, Odds ratio (OR) in a cohort study. B, Odds ratio (OR) in a case-control study.
Downloaded from: StudentConsult (on 8 October 2009 11:44 AM) © 2005 Elsevier
Prevalence Odds Ratio (POR) = Cross Product Ratio bila data didasarkan pada kasus-kasus prevalens Prevalence Odds Ratio
650 x 350 950 x 50
Prevalence ( proportion) Ratio
650 / 1600 50 / 400
4,8 0,40625 0,125
Faktor Perokok
Kasus 650 (a)
Kontrol 950 (b)
Total 1600
Bukan perokok Total
50 (c)
350 (d)
400
700
1300
2000
3,25
Odds Ratio & Risk Ratio The odds ratio will provide a good estimate of the risk ratio when: 1. The outcome (disease) is rare E+ E-
D+ a c
Db d
OR = (a / c) / (b / d) OR = (ad) / (bc)
a / (a +b ) RR = -----------c / (c +d)
If the disease is rare, then cells (a) and (c) will be small a / (a +b ) a / b ad RR = ------------ = ------ =-- = OR c / (c +d) c / d bc
Figure 11-6 Example: The odds ratio is a good estimate of the relative risk when a disease is infrequent. Downloaded from: StudentConsult (on 8 October 2009 11:44 AM) © 2005 Elsevier
Figure 11-7 Example: The odds ratio is not a good estimate of the relative risk when a disease is not infrequent. Downloaded from: StudentConsult (on 8 October 2009 11:44 AM) © 2005 Elsevier
In a prospective study/cohort study, the Relative Risk can be calculated directly In a retrospective study /case control study , the RR cannot be calculated directly, so that the Relative Odds or ODDS RATIO ( Cross Products Ratio ) is used as an estimate of the RR, when the risk of the disease is low
OR & RR • Pada penyakit yang jarang terjadi,nilai Odds Ratio hampir sama dengan nilai Relative Risk (Risk Ratio). Nilai Prevalence Odds Ratio hampir sama dengan nilai Prevalence Proportion Ratio. • Pada penyakit yang umum terjadi, nilai Odds Ratio lebih ekstrim dari pada Risk Ratio.
Interpretation of Odds Ratio • Relative odds associated with exposure OR = 1 OR > 1 OR < 1
no association positive association negative association
• Size of OR indicates strength of association • OR ≈ RR when disease rare (i.e., risk < 5%); when disease not rare, OR still a valid measure of association
Gerstman
Chapter 8 (partial)
28
