| Abstract | Hipoteza ove disertacije bila je da je šećerna bolest slabo prepoznata kao uzrok smrti
te se spominje kao jedan od višestrukih uzroka smrti u tri puta više umrlih osoba
nego što je navedeno u službenim statistikama koje se zasnivaju samo na osnovnom
uzroku smrti. Također, hipoteza je bila da se u umrlih osoba s prethodno
dijagnosticiranom šećernom bolešću ona rijetko navodi kao uzrok smrti te je
prepoznata kao jedan od višestrukih uzroka smrti u manje od polovice umrlih osoba s
dijabetesom u Hrvatskoj.
Ciljevi disertacije su bili odrediti prepoznatost dijabetesa na potvrdama o smrti u
općoj populaciji kao osnovnog uzroka smrti i jednog od višestrukih uzroka smrti;
istražiti uzroke smrti i prepoznatost šećerne bolesti kao uzroka smrti u osoba s
prethodno poznatom šećernom bolešću (iz CroDiab registra) u Hrvatskoj; te odrediti
čimbenike koji su povezani s češćim navođenjem dijabetesa kao jednog od uzroka
smrti u općoj populaciji i u populaciji osoba sa šećernom bolešću zabilježenih u
CroDiab registru.
U prvom dijelu istraživanja pregledane su potvrde o smrti za sve preminule osobe u
Hrvatskoj 2010. godine (N=52.096). Na temelju tih podataka su određeni mortalitetni
udjeli šećerne bolesti kao osnovnog uzroka smrti i kao jednog od višestrukih uzroka
smrti. U drugom dijelu istraživanja su ovi podaci upareni s podacima o kohorti osoba
sa šećernom bolešću iz CroDiab registra (N=66.329). Tako su utvrđeni uzroci smrti u
osoba sa šećernom bolešću preminulih 2010. godine (N=2.593, gruba stopa
mortaliteta 3.909/100.000), udjeli prepoznatog dijabetesa na potvrdama o smrti u
ovih bolesnika te čimbenici povezani s navođenjem dijabetesa na potvrdama o smrti
u bolesnika s poznatom šećernom bolešću. Podaci o zastupljenosti dijabetesa na
potvrdama o smrti i uzroci smrti prikazani su metodama deskriptivne statistike, dok su
za predikciju pojavnosti dijabetesa na potvrdama o smrti i određivanje čimbenika
rizika korišteni bivarijatni i multivarijatni modeli logističke regresije.
U općoj populaciji je šećerna bolest bila osnovni uzrok smrti u 3% umrlih žena i 2,5%
umrlih muškaraca (p<0,001), a bila je jedan od višestrukih uzroka smrti u 12,3%
umrlih žena i 10,6% umrlih muškaraca (p<0,001). Šansa za pojavu dijabetesa na
potvrdi o smrti je bila 7% veća kod žena, 13% veća ukoliko je smrt nastupila u
zdravstvenoj ustanovi, te 51% manja ukoliko je uzrok smrti odredio obducent.
Dijabetes je bio osnovni uzrok smrti u 14,9% preminulih osoba sa šećernom bolešću
(bez razlika po spolu), a najčešći uzrok smrti u ovoj skupini bolesnika su bile
kardiovaskularne bolesti (48,4%). Među svim preminulim osobama iz CroDiab
registra, šećerna bolest je bila navedena u 51,8% potvrda o smrti (bez razlika po
spolu). U multivarijatnom modelu značajni prediktori su bili određivanje uzroka smrti
od strane obducenta (60% manja šansa) i trajanje dijabetesa duže od 15 godina
(89% veća šansa).
Zaključno, šećerna bolest je prepoznata kao osnovni uzrok smrti svake sedme
oboljele osobe te je prepoznata kao jedan od višestrukih uzroka tek u nešto više od
polovice svih smrtnih ishoda u osoba sa šećernom bolešću. U skladu s time,
oslanjanje na podatke službene mortalitetne statistike značajno podcjenjuje
opterećenje bolešću uzrokovano dijabetesom u Hrvatskoj, što je potrebno uzeti u
obzir pri planiranju intervencija na razini populacije.
Službena mortalitetna statistika u Hrvatskoj se još uvijek zasniva samo na osnovnim
uzrocima smrti, što podcjenjuje javnozdravstveni značaj bolesti koje su često
navedene kao jedan od uzroka smrti, poput šećerne bolesti. Zbog ovoga bi se trebalo
razmisliti o uvođenju sustava mortalitetne statistike koji bilježi više uzroka smrti.
Takav sustav, koji je u funkciji u velikom broju europskih država, omogućuje bolji uvid
u javnozdravstveni značaj i mortalitetni udio pripisiv pojedinim bolestima, uključujući i
šećernu bolest. |
| Abstract (english) | The hypothesis of this dissertation was that diabetes mellitus is insufficiently
recognized as a cause of death and that, in deceased persons in Croatia, it is
mentioned on death certificates three times more often than it is coded as an
underlying cause of death. It was also hypothesized that, in deceased persons with
previously diagnosed diabetes mellitus, diabetes is rarely mentioned as an underlying
cause of death and that it is mentioned on less than a half of those death certificates.
The aims of this dissertation were to establish the recognition of diabetes on death
certificates in the general population both as the underlying cause of death and one
of the multiple causes of death; to explore the causes of death in persons with
diagnosed diabetes mellitus (from the CroDiab registry) in Croatia; and to establish
the factors that are associated with more frequent citing of diabetes as one of the
causes of death, both in the general population and in the population of persons with
diabetes from the CroDiab registry.
In the first part of the study, death certificates for all deceased persons in Croatia for
the year 2010 were examined (N=52,096). Based on this data, mortality fractions for
diabetes as underlying cause of death and as one of multiple causes of death were
established. In the second part of the study, these data were combined with data on
the cohort of persons with diabetes from the CroDiab registry (N=66,329). The
causes of death for persons with diabetes who had deceased in 2010 were
established (N=2,593, crude mortality rate 3,909/100,000), as were the proportion of
death certificates mentioning diabetes in these patients, and the factors associated
with citing diabetes on these death certificates. The data on the frequency of
reporting of diabetes on death certificates and the causes of death were presented
by descriptive statistics. The prediction of citing of diabetes on death certificates and
the factors of risk were determined by bivariate and multivariate logistic regression
models.
In the general population, diabetes was recognized as the underlying cause of death
in 3% of deceased women and 2.5% of deceased men (p<0.001), and as one of
multiple causes of death in 12.3% of deceased women and 10.6% of deceased men
(p<0.001). The odds of diabetes appearing on the death certificate were 7% higher in
women, 13% higher if death had occurred in a health institution, and 51% lower if the
cause of death was determined by a pathologist.
Diabetes was the underlying cause of death in 14.9% of deceased persons with
diabetes mellitus (without sex differences), and the most common cause of death in
these persons were cardiovascular diseases (48.4%). Among all deceased persons
with diabetes from the CroDiab registry, diabetes was cited on 51.8% of death
certificates (without sex differences). In the multivariate model, significant predictors
were the cause of death assigned by a pathologist (60% lesser odds) and diabetes
duration longer than 15 years (89% greater odds).
In conclusion, diabetes is recognized as the underlying cause of death in one in
seven persons with diabetes, and it is recognized as one of multiple causes of death
in just above one half of all deaths in persons with diabetes. Therefore, the reliance
on official mortality statistics significantly underestimates the burden of diabetes in
Croatia, which should be taken into consideration when planning population-based
public health interventions.
The official mortality statistics in Croatia is still based on the underlying causes of
death only, which significantly underestimates the burden of disease associated with
certain diseases that are often mentioned as one of the multiple causes of death,
such as diabetes. Therefore, the implementation of mortality statistics coding system
which codes multiple causes of death should be considered. This system, operating
in many European countries, provides a better insight in the public health importance
and mortality fractions attributable to various diseases, including diabetes. |
