Incidence of IDDM during 1988–1992 in Zagreb, Croatia

Diabetologia, May 1995

Summary The objective of this study was to determine the incidence of insulin-dependent diabetes mellitus (IDDM) in the population of Zagreb, Croatia, during 1988–1992. A centralized diabetes registry was the primary source of data, while secondary sources were used to assess ascertainment. A total of 282 new cases of IDDM were diagnosed in the study period, the primary and secondary sources identifying annually 93–100% of the cases. The annual incidence rate ranged from 5.6 per 100,000 to 6.6 per 100,000. Early fatality in persons older than 50 years was the major cause of underascertainment. The incidence peaked in the 10–14 years age group (12.4 per 100,000), and remained stable after age 24 years. Males had a significantly higher incidence in the 5–9 and 24–44 years age groups. In the 45–54 years age group, females had a significantly higher incidence. No seasonality was observed. Despite the war conditions in Croatia, the low overall IDDM incidence rates did not change significantly during the study period.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://link.springer.com/content/pdf/10.1007%2FBF00400723.pdf

Incidence of IDDM during 1988–1992 in Zagreb, Croatia

Diabetologia Incidence of IDDM during 1 9 8 8 - 1 9 9 2 in Zagreb, Croatia 0 1The Vuk Vrhovac Institute for Diabetes, Endocrinology and Metabolic Diseases , Zagreb , Croatia 2Department of Pediatrics, Clinical Hospital "Sestre Milosrdnice", Zagreb, Croatia 3Department of Pediatrics, Clinical Hospital "Rebro" , Zagreb , Croatia G. R o g l i ~ 1, 1. P a v l i ~ - R e n a r 1, S. S e s t a n - C r n e k 1, M. P r a g e k 1, M. K a d r n k a - L o v r e n ~ i ~ a, A . R a d i c a 3, Z. M e t e l k o t Summary The objective of this study was to determine the incidence of insulin-dependent diabetes mellitus ( I D D M ) in the population of Zagreb, Croatia, during 1988-1992. A centralized diabetes registry was the primary source of data, while secondary sources were used to assess ascertainment. A total of 282 n e w cases of I D D M were diagnosed in the study period, the primary and secondary sources identifying annually 93-100 % of the cases. The annual incidence rate ranged from 5.6 per 100,000 to 6.6 per 100,000. Early fatality in persons older than 50 years was the major cause of underascertainment. The incidence p e a k e d in the 10-14 years age group (12.4 per 100,000), and remained stable after age 24years. Males had a significantly higher incidence in the 5-9 and 24-44 years age groups. In the 45-54 years age group, females had a significantly higher incidence. No seasonality was observed. Despite the war conditions in Croatia, the low overall I D D M incidence rates did not change significantly during the study period. [Diabetologia (1995) 38: 550-554] Insulin-dependent epidemiology; war - 9 Springer-Verlag 1995 Since the early 1980s an increasing b o d y of knowledge h a s b e e n accumulated in a standardised manner on the incidence of insulin-dependent diabetes mellitus ( I D D M ) throughout the world [ 1, 2 ]. H o w ever, these standardised incidence registries mostly provide data for the 0-14 years age group. Only a few studies have provided data on the incidence of I D D M in persons older than 30 years, with somewhat different definitions of I D D M [ 3-8 ]. To our knowledge, there are no reports of population-based studies of I D D M incidence in persons older than 30 years in a population of Slavic origin. In Croatia registration of diabetic patients has b e e n compulsory since 1975 [ 9 ]. Previous research has indicated that the incidence of childhood diabetes in the population of Zagreb, Croatia was low during 1970-1989 [ 10 ]. This study was undertaken to determine the incidence of I D D M in the total population of Zagreb during 1988-1992, and to test the hypothesis that the war in Croatia affected the incidence of I D D M in the residents of Zagreb. Stress has b e e n suspected to play a role in the onset of I D D M . The literature on psychoimmunology suggests several mechanisms by which psychological stress may be implicated in the development of I D D M [ 11 ], but conclusive evidence is lacking. Subjectsand methods Zagreb, the capital of Croatia and its largest city, covers an area of 1,705 square kin, the population size being 933,914 in 1991, which is about 20 % of the total Croatian population. Base population estimates were obtained from the Central Bureau of Statistics of Croatia [ 12 ], based on decennial 1981 and 1991 population censuses, and intercensual estimates for 1988, 1989, 1990, and 1992. All persons suspected of having diabetes in Croatia are referred to a diabetologist for diagnosis and initial treatment. In Zagreb, all suspected cases are referred to one of the two pediatric diabetes centres if the perG. Roglid et al.: Incidence of I D D M in Croatia son is less than 15 years old, or to the Institute for Diabetes. All persons hospitalized for any reason are referred to a diabetologist for follow-up if diabetes is diagnosed during hospital treatment. Upon diagnosis all persons are registered in the central diabetes registry. Insulin syringes, self-monitoring devices and test-strips can only be obtained free of charge if prescribed by a diabetologist. There are no private hospitals or private diabetologists. The diabetes registry was the primary source of IDDM incidence data. In order to include a person in the calculation of I D D M incidence rates, he or she had to: 1) have diabetes diagnosed after 31 December 1987 and before 1 January 1993; 2) reside in the administrative area of Zagreb at the time of diagnosis; 3) present with laboratory evidence of ketonuria or ketoacidosis at the time of diagnosis and/or be placed on insulin treatment within i week of diagnosis; 4) remain on continuous insulin treatment, allowing for a brief "honeymoon" period in the first year of treatment. All case records of persons registered in the study period were examined. To ensure as complete a case ascertainment as possible and to estimate the actual number of cases, two additional independent data sources were used together with the central diabetes registry. One of them consisted of the death certificates of all deceased in the Zagreb area during 1988-1991. These death certificates were checked to identify persons that had been diagnosed as having diabetes and who died before being registered. All certificates with the diagnosis of diabetes mellitus were cross-referenced with the registry data. The medical records of deceased persons not found in the registry were reviewed to determine the type of diabetes. The Association of Diabetic Patients is an independent charity organization which organizes free educational holidays for diabetic children, distributes free food packages and provides other kinds of material aid to its members. The membership files were reviewed for persons on insulin treatment, diagnosed 1988-1992, and the identified persons' data were checked for registration in the central diabetes registry. Refugees and displaced persons living in Zagreb at the time of diagnosis were not included in the numerators for incidence calculation because of the uncertain number of all refugees and displaced persons in Zagreb. Statistical analysis Completeness of ascertainment was estimated by the twosource capture-recapture method, merging all data from secondary sources [ 13 ]. The observed numbers of cases in each age group and calendar year were used in the calculation of incidence rates [ 14 ]. The incidence rates are expressed as age-and sex-specific rates per 100,000 population, and rates for the total population are expressed as age-adjusted rates per 100,000 population, using the direct method and the total population of Croatia as the standard population [ 12, 15 ]. Denominators of incidence rates over 1988-1992 are the sum of the denominators of each year. Assuming a Poisson distribution of the observed number of cases, the 95 % confidence intervals (Cls) were computed [16]. Statistical analysis of temporal trend was performed by linear regression, using SAS statistical software. A non-parametric procedure was used for testing for seasonal variation in diagnosis [ 17 ]. Seasons were defined in 3-month windows beginning with January. Results I n t h e p e r i o d 1988-1992, 282 i n c i d e n t cases o f I D D M w e r e i d e n t i f i e d f r o m a m o n g t h e r e s i d e n t s o f Z a g r e b . P e r s o n s w i t h s e c o n d a r y d i a b e t e s w e r e n o t i n c l u d e d . T h e c e n t r a l d i a b e t e s r e g i s t r y i d e n t i f i e d 274 cases. T h e d a t a f o r e s t i m a t i o n o f c o m p l e t e n e s s a r e p r e s e n t e d in T a b l e 1. A l l p e r s o n s i d e n t i f i e d t h r o u g h the A s s o c i a t i o n o f D i a b e t i c P a t i e n t s h a d p r e v i o u s l y b e e n i d e n t i f i e d t h r o u g h the p r i m a r y source. D e a t h certific a t e s i d e n t i f i e d e i g h t p e r s o n s w h o e s c a p e d r e g i s t r a t i o n b e c a u s e o f d e a t h o n diagnosis: t w o m a l e s ( a g e d 62 a n d 63 y e a r s in 1988, o n e f e m a l e a g e d 66 y e a r s in 1990, a m a l e a g e d 61 a n d a f e m a l e a g e d 69 y e a r s in 1991, t w o f e m a l e s ( a g e d 54 a n d 26 y e a r s ) a n d o n e a Significant difference, p < 0.05 Person-years male aged 52 years in 1992. Thus, the primary and secondary sources identified at least 85 % and probably m o r e of newly-diagnosed patients in the study period. Except for the 26-year-old female, underascertainment was restricted to the over-50 y e a r age group. T a b l e 2 shows age- and sex-specific rates for I D D M for the period 1988-1992. The incidence of I D D M was lowest in children aged 0-4 years, significantly increasing to a peak at age 10-14 years. In the whole group, the peak incidence declines significantly in the 15-24-year age group, declining further after age 24 years, and remaining stable in older age groups. Sex- and age-specific rates reveal a significantly higher incidence in males than in females in the 5-9 and 25--44 year age groups (p < 0.05), the decline after age 24 years being less pronounced in males than in females, and non-significant until after age 44 years. However, females had a significantly higher rate in persons aged 45-54 years (p < 0.05). No significant temporal differences (linear trend or deviation from m e a n distribution) were observed during the study period in the whole group, nor were t h e y detected in different sex (Table 3) or age groups. Although a greater proportion of cases was diagnosed in a u t u m n and winter, an uneven seasonal distribution of I D D M cases could not be confirmed statistically for either the whole group, or for age groups 0-14 years, 15-34 years and 35+ years (p > 0.05). The seasonal pattern was not different for males or females (p > 0.05). Discussion The definition of I D D M in this study is a possible cause of underascertainment. Melton et al. [ 4 ] have shown that m o r e than 90 % of diabetic patients starting insulin t r e a t m e n t within 1 year of diagnosis are started on insulin during the first week after diagnosis. Still, a n u m b e r of patients progressing to insulin therapy after some months have not been included in the numerators. On the other hand, our data might include "false positive" I D D M patients. Although it has been shown that m o r e than 80 % of withdrawals from insulin occurred during the first year [ 4 ], the decision to withdraw insulin could depend on the individual physician. Also, some patients were followed for less than a year after diagnosis. The primary source identified a high proportion of patients, the ascertainment varying only slightly in the study period. However, the existing diabetes registry does not seem to be adequate for identifying incident cases in persons older than 50 years. Early fatality seems to influence the incidence rates in this age group, since 14 % of incident cases in this age group was identified solely through death certificates, thus supporting data on the high mortality of newly-diagnosed I D D M in the elderly [ 18 ]. Of the patients who died before being registered, none had had diabetes diagnosed previously. Two patients died at home, without medical attendance. The coroner ascribed their death to diabetic ketoacidosis (DKA). The other patients were admitted to hospital in D K A and died within 2-48 h of admission. Only one patient was conscious on admission. In at least four of the eight deceased patients, death could be ascribed exclusively to D K A , while the others subsequently developed conditions which probably contributed to early death. It was not possible to determine how m a n y patients had sought medical care prior to admission with D K A . Therefore it is difficult to say w h e t h e r this serious condition on admission to hospital is a result of failure to recognize the symptoms of D K A in middle-aged and elderly persons, or a result of swift metabolic deterioration without pron o u n c e d symptoms. The incidence rates in this study are lower than those reported for several other countries in a similar study period, particularly in persons younger than 30 years [ 6, 8, 19, 20 ]. Childhood incidence of I D D M in Zagreb appears to be low and similar to that of other E u r o p e a n populations of Slavic origin and populations of Southern E u r o p e [2]. The peak incidence occurring in the 10-14 year age group in this Croatian population has also been reported for m a n y other populations with very different rates [ 20-24 ]. The rates in the 15-30 age group are closest to those found in Turin and the Turin Province, and the L o m b a r d y region in Italy [ 21, 22, 25, 26 ]. It has been proposed to categorize populations into low, intermediate and high I D D M risk groups, according to incidence in persons younger than 20 years [ 1 ]. If these criteria are accepted, the population of Zagreb is not at high risk of I D D M , the incidence being less than 15 per 100,000. There are only a few studies describing the incidence of I D D M in all age groups, and most research has been done in high-risk populations [ 4, 8, 19, 27 ]. The stable incidence after a decline b e y o n d age 24 years is a finding different from the decrease in incidence after age 50 years observed in Estonia [19]. It is also different f r o m the findings of a second peak occurring in old age [ 4, 8, 28 ]. A higher incidence in males was f o u n d in the 5-9year-old children, but not in teenagers and adults y o u n g e r t h a n 25 years, as f o u n d in some studies [ 6, 20, 21, 23 ]. However, in the 25-44 year age group the incidence of I D D M is significantly higher in males, thus supporting the hypothesis of the possible effect of lifestyle, rather t h a n sex hormones, on I D D M incidence [6]. However, this difference in risk does not persist in older age groups. A higher incidence in females was f o u n d in the age group 45-54 years, and b o t h sexes had a similar incidence after age 55 years. This finding is difficult to explain in terms of lifestyle factors. No secular trend was detected in the study period. The small size of the study population is likely to have decreased the power of detecting temporal variation. We had hypothesized that the war-related events in Croatia affected the incidence of I D D M in the population of Zagreb. During the war in Croatia, starting in the s u m m e r of 1991, the area of Zagreb was not the site of direct combat. However, the city was b o m b e d once, there were daily air-raid alarms and the sounds of combat were audible from the front line only 30 kilometres away. M a n y inhabitants, particularly y o u n g e r males, were directly involved in warfare. The incidence of I D D M does not seem to have b e e n affected by war-related events. Although it was expected that an effect could be demonstrated within 18 months of exposure to war-related stressors, it is too early to conclude that the effect of war was negligible, as clinical onset of I D D M can be slow [ 29 ]. The absence of seasonal differences in the proportion of cases diagnosed could be caused by the relatively small sample size, although no seasonality had previously b e e n r e p o r t e d from other studies conducted only in younger age groups [ 21, 30, 31 ]. The risk of I D D M in this population of Slavic origin is low, and the rates were not affected by the war so far. Underascertainment due to early fatality in older ages is likely to alter the I D D M rates. It is possible that this is a local, population-specific finding, therefore studies in other populations are necessary. It is recomm e n d e d that death certificates be included as means of case ascertainment w h e n measuring I D D M incidence in persons over 50 years of age, unless it has been d e m o n s t r a t e d that this source does not improve the completeness of data. Acknowledgements.Part of this study was performed through collaboration with the EU Concerted Action Project EURODIAB ACE (Contract No. BMH1 CT92-0043). 1. LaPorte ER , Tajima N , Akerblom HK et al. ( 1985 ) Geographic differences in the risk of insulin-dependent diabetes mellitus: the importance of registries . Diabetes Care 8 [Suppl 1 ]: 101 - 107 2. Green A , Gale EAM , Patterson CC , The EURODIAB Subarea A Study Group ( 1992 ) Incidence of childhood-onset insulin-dependent diabetes mellitus: The EURODIAB ACE Study . Lancet 339 : 905 - 909 3. Laakso M , Py6r~il~iK ( 1985 ) Age of onset and type of diabetes . Diabetes Care 8 : 114 - 117 4. Melton LJ , III, Palumbo PJ , Chu-Pin C ( 1983 ) Incidence of diabetes mellitus by clinical type , Diabetes Care 6 : 75 - 86 5. Michaelis D , Jutzi E , Vogt L ( 1993 ) Epidemiology of insulin-treated diabetes mellitus in the East-German population: differences in long-term trends between incidence and prevalence rates . Diabete Metab 19 : 110 - 115 6. Blohm6 G , Nystrtim L , Arnqvist HJ et al. ( 1992 ) Male predominance of type 1 (insulin-dependent) diabetes mellitus in young adults: results from a 5-year prospective nationwide study of the 15-34-year age group in Sweden . Diabetologia 35 : 56 - 62 7. Pagano G , Cavallo-Perin R Cavalot F et al. ( 1987 ) Genetic, immunologic, and environmental heterogeneity of IDDM. Incidence and 12-mo follow-up of an Italian population . Diabetes 36 : 859 - 863 8. Scott RS , Brown LJ ( 1991 ) Prevalence and incidence of insulin-treated diabetes mellitus in adults in Canterbury . New Zealand. Diabet Med 8 : 443 - 447 9. Skrabalo Z , Metelko Z , Aganovi6 I , Granid M ( 1987 ) Croatian Model - to consider appropriate technology for diabetes care programs in the community . Diabetologia Croatica 16 : 109 - 140 10. Roglid G , Metelko Z , Radica A , Kadrnka-Lovren~id M , Grani6 M , Skrabalo Z ( 1991 ) Incidence of insulin-dependent diabetes mellitus - the Zagreb incidence registry . Diabetes 40 [Suppl 1 ]: 319 (Abstract) 11. Solomon GF , Amkraut AA , Rubin RT ( 1985 ) Stress, hormones, neuroregulation and immunity . In: Burchfield SR (ed) Stress: psychological and physiological interactions . Hemisphere, Washington, pp 207 - 221 12. Popis stanovnigtva 1991 . Stanovnigtvo prema spolu i starosti . Republika Hrvatska . (Population census. Population by sex and age . Republic of Croatia) . Dr~avni zavod za statistiku . Dokumentacija 882 . Zagreb, 1994 13. Wittes JT , Colton T , Sidel VW ( 1974 ) Capture-recapture methods for assessing the completeness of case ascertainment when using multiple information sources . J Chron Dis 27 : 25 - 36 14. LaPorte RE , McCarty D , Bruno G , Tajima N , Baba S ( 1993 ) Counting diabetes in the next millenium . Diabetes Care 16 : 528 - 534 15. Lilienfeld AM , Lilienfeld DE ( 1980 ) Foundations of epidemiology .Oxford University Press, New York Oxford, p 353 16. Lilienfeld AM , Lilienfeld DE ( 1980 ) Foundations of epidemiology . Oxford University Press, New York Oxford, pp 336 - 338 17. Freedman LS ( 1978 ) The use of a Kolmogorov-Smirnov type statistic in testing hypotheses about seasonal variation . J Epidemiol Community Health 33 : 525 - 538 18. Gale EAM , Dornan TL , Tattersall RB ( 1981 ) Severely uncontrolled diabetes in the over fifties . Diabetologia 21 : 25 - 28 19. Kalits I , Podar T ( 1990 ) Incidence and prevalence of type 1 (insulin-dependent) diabetes in Estonia in 1988 . Diabetologia 33 : 346 - 349 20. Goday A , Castell C , Tresserras R , Canela J , Taberner JL , Lloveras G and the Catalan Epidemiology Diabetes Study Group ( 1992 ) Incidence of type i (insulin-dependent) diabetes mellitus in Catalonia, Spain . Diabetologia 35 : 267 - 271 21. Bruno G , Merletti F , Vuolo A , Pisu E , Giorio M , Pagano G ( 1993 ) Sex differences in incidence of IDDM in age group 15-29 yr. Higher risk in males in Province of Turin, Italy . Diabetes Care 16 : 133 - 136 22. Bruno G , Merletti F , Pisu E , Pastore G , Marengo C , Pagano G ( 1990 ) Incidence of IDDM during 1984-1986 in population aged < 30 yr residents of Turin, Italy . Diabetes Care 13 : 1051 - 1056 23. Allen C , Palta M , D'Alessio DJ ( 1986 ) Incidence and differences in urban-rural seasonal variation of type 1 (insulin-dependent) diabetes in Wisconsin . Diabetologia 29 : 629 - 633 24. Soltesz G , Madacsy L , Bekefi D , Danko I , the Hungarian Childhood Diabetes Epidemiology Group ( 1989 ) Rising incidence of type 1 diabetes in Hungarian children ( 1987 - 1987 ). Diabetic Med 7 : 111 - 114 25. Garancini P , Gallus G , Calori G , Formigaro F , Micossi P ( 1991 ) Incidence and prevalence rates of diabetes mellitus in Italy from routine data: a methodological assessment . Eur J Epidemiol 7 : 55 - 63 26. Calori G , Gallus G , Garancini P , Repetto F , Micossi P ( 1990 ) Identification of the cohort of type I diabetes presenting in Lombardy in 1983-1984: a validated assessment . Diabet Med 7 : 595 - 599 27. Christau B , Mr AG ( 1984 ) Further epidemiological evidence of a higher incidence level for the insulin-dependent diabetes mellitus in older age groups . Acta Endocrinol [Suppl] 95:18 28. Krolewski AS , Warram JH , Rand LI , Kahn CR ( 1987 ) Epidemiologic approach to the etiology of type 1 diabetes mellitus and its complications . N Engl J Med 317 : 1390 - 1398 29. Tarn AC , Smith CP , Spencer KM , Bottazzo GF , Gale EAM ( 1987 ) Type I (insulin-dependent) diabetes: a disease of slow clinical onset? B M J 294 : 342 - 345 30. Lorenzi M , Cagliero E , Schmidt NJ ( 1985 ) Racial differences in incidence of juvenile-onset type 1 diabetes: epidemiologic studies in southern California . Diabetologia 28 : 734 - 738 31. Crossley JR , Upsdell M ( 1980 ) The incidence of juvenile diabetes mellitus in New Zealand . Diabetologia 18 : 29 - 34


This is a preview of a remote PDF: https://link.springer.com/content/pdf/10.1007%2FBF00400723.pdf

G. Roglić, I. Pavlić-Renar, S. šestan-Crnek, M. Prašek, M. Kadrnka-Lovrenčić, A. Radica, Ž. Metelko. Incidence of IDDM during 1988–1992 in Zagreb, Croatia, Diabetologia, 1995, 550-554, DOI: 10.1007/BF00400723