| Sažetak | Ciljevi istraživanja:
Cilj ovog istraživanja bio je generirati podatke za KEP-indeks čitave studentske populacije
Sveučilišta u Splitu iz tri uzorka različite veličine korištenjem Monte Carlo metode.
Materijali i metode:
Podatci za potrebe istraživanja prikupljeni su u sklopu kliničkih vježbi na kolegijima
„Restaurativna dentalna medicina 2“ te „Endodoncija 2“. Svi ispitanici (n =200) su bili studenti
Sveučilišta u Splitu koji su došli na pregled ili liječenje u ordinacije fakultetske poliklinike za
dentalnu medicinu „Dental academicus“. Iz prikupljenih podataka o KEP-indeksu izvedena su tri
uzorka – mali uzorak (n =50), uzorak srednje veličine (n = 100) i cjelokupni uzorak (n = 200). Iz
navedenih uzoraka Monte Carlo simulacijama (MCS) su generirani podatci o KEP-indeksu za tri
studentske populacije (MCS 50, MCS 100 i MCS 200) od kojih je svaka bila veličine 20 000
pojedinaca. Vjerojatnosti pojedinih ishoda za KEP-indeks i njegove sastavnice, kao i međusobna
povezanost KEP-indeksa i sastavnica KEP-indeksa su ubačeni kao pretpostavke za MCS.
Rezultati:
Procjena za prosječni KEP-indeks studentske populacije prema MCS 50 je 8,96 ± 0,69 (99% CI
[8,91; 9]), prema MCS 100 9,12 ± 0,47 (99% CI [9,08; 9,21]), te prema MCS 200 8,82 ± 0,36
(99% CI [8,77; 8,87]). Od sastavnica KEP-indeksa, broj saniranih zubi (P) je u sve tri simulacije
bio najjače povezan s KEP-indeksom i to R = 0,84 (MCS 50), R = 0,82 (MCS 100) i R = 0,77
(MCS 200). Procjene prosječnog KEP-indeksa MCS-om odstupale su za manje od 1 boda u odnosu
na procjene prosječnog KEP-indeksa iz pripadajućih uzoraka i međusobno.
Zaključci:
Monte Carlo metoda može biti korisna u procjeni prosječnih populacijskih vrijednosti kliničkih
indeksa u dentalnoj medicini uključujući i KEP-indeks. Prema postavkama MCS u ovom
istraživanju, veličina izvornih uzoraka nije bitnije utjecala na završnu procjenu parametara KEPindeksa. |
| Sažetak (engleski) | Objectives:
The aim of this study was to generate data for the DMFT (Decayed, Missing, and Filled Permanent
Teeth) index of the entire student population of the University of Split from three samples of
different sizes using Monte Carlo method.
Materials and methods:
Data for this study were collected during clinical exercises in the courses "Restorative Dental
Medicine 2" and "Endodontics 2". All participants (n = 200) were students of the University of
Split who came for an examination or treatment in the offices of the faculty polyclinic for dental
medicine "Dental academicus". Three samples were derived from the collected data on the DMFT
index - a small sample (n = 50), a medium-sized sample (n = 100) and the entire sample (n = 200).
Monte Carlo simulations (MCS) were used to obtain DMFT index data for three student
populations (MCS 50, MCS 100, and MCS 200) of 20,000 individuals each from the above
samples. The probabilities of individual outcomes for the DMFT index and its components, as well
as the mutual correlation between the DMFT index and the components of the DMFT index, were
entered as assumptions for the MCS.
Results:
The estimated mean DMFT index of the student population by MCS 50 was 8.96 ± 0.69 (99% CI
[8.91; 9]), by MCS 100 was 9.12 ± 0.47 (99% CI [9.08; 9.21]), and by MCS 200 was 8.82 ± 0.36
(99% CI [8.77; 8.87]). Regarding the components of the DMFT index, the number of repaired
teeth (F) in MCS 50, MCS 100, and MCS 200 was most strongly correlated with the DMFT index,
namely R = 0.84, R = 0.82, and R = 0.77, respectively. The estimated mean DMFT indices by
MCS differed from the mean DMFT indices of the corresponding samples and from each other by
less than 1 point.
Conclusions:
The Monte Carlo method may be useful in estimating the population means of clinical indices in
dental medicine, including the DMFT index. According to the assumptions made for MCS in this study, the size of the original samples did not significantly affect the final estimate of the
parameters of the DMFT index. |
