Downloaded 58 times
Uploaded byBasil Tumaini
Standardization of rates by Dr. Basil Tumaini
The document discusses the concept of rate standardization, specifically in epidemiology, covering definitions, types of rates, and methods for direct and indirect standardization. It emphasizes the importance of adjusting rates to account for population heterogeneity, especially age distribution, in order to facilitate unbiased comparisons. The document also details procedures for calculating crude, specific, and standardized rates, illustrating the implications of these rates with examples.
In this document
Powered by AI
Introduces the presentation by Dr. Basil Tumaini on rate standardization, covering definitions, types of rates, and the role of confounding.
Defines rate as a measure of event occurrence over time in populations, highlighting examples and calculations, such as the mortality rate in Arusha.
Covers crude, specific, and adjusted rates. Details advantages and disadvantages of crude rates in reflecting population heterogeneity.
Discusses specific rates that categorize populations into subgroups based on demographics, including examples of infant mortality and age-specific rates.
Presents age-specific TB incidence rates among different population groups, comparing native and immigrant populations using graphical data.
Explains issues with crude rates and emphasizes the necessity for standardizing rates for accurate population comparisons based on demographic variables.
Describes direct standardization procedures, requiring specific rates from populations, and discusses how expected deaths are calculated.
Explains the process of concluding expected cases/deaths and deriving standardized rates based on community data.
Outlines the use of indirect standardization for populations with insufficient data, detailing the standardized incidence ratio calculations.
Provides comprehensive procedures for indirect standardization, calculating standardized mortality ratios and their implications.
More Related Content
Similar to Standardization of rates by Dr. Basil Tumaini
More from Basil Tumaini
Standardization of rates by Dr. Basil Tumaini
- 1. Rate Standardization Dr. Basil Tumaini MMEDResident Muhimbili University of Health and Allied Sciences
- 2. Outline Definition ofterms Types of rates The role of confounding and the rationale for standardisation Direct and indirect standardisation Discussion
- 3. Rate Defined asspecial form of proportion that includes a specification of time. Most commonly used in epidemiology because it most clearly expresses probability or risk of diseases or other events in defined population over a specified period of time.
- 4. Calculation Of Rates Rate= Number of events in a specified time period Population at risk of these events in a specified time period X k k is used to denote the units of population such as per 1,000 or per 100,000 depending on convention and convenience 9,981 deaths in Arusha in 2000 951,270 total population in Arusha 2000 = 10.49 per 1,000 or 1049 per 100,000
- 5. Types of Rates 3 major types: Crude rates Specific rates (age-specific, e.g., infant mortality rate) Adjusted rates
- 6. Crude Rates Aresummary statistics that ignore heterogeneity of the population under investigation Example: Crude mortality rate: Total deaths in 2003 x 1,000 = T.Z. death rate Estimated T.Z. pop in 2003*
- 7. Crude Rates Number ofnew cases in a specified year ---------------------------------- x 1000 Number of individuals in the population in the specified year Number of deaths in a specified year -------------------------------- x 1000 Number of individuals in the population in the specified year Crude mortality rate = Crude incidence rate =
- 8. Crude Rates Advantages • ActualSummary rates • Easy calculation for international comparisons Disadvantages Since population varies in composition (e.g., age) differences in crude rates difficult to interpret
- 9. Specific Rates Stratifies populationsinto more homogeneous groups (strata) based on the demographic characteristic thought to be related to the outcome of interest (e.g. age-specific, sex- specific, race-specific)
- 10. Example Total # ofdeaths in 2003 among persons age less than 1 year x 1,000 = 2003 infant mortality rate Number of live births during 2003
- 11. Age-Specific Mortality Rate •Provides a broader view of mortality for sub-groups stratified by age • Numerator and denominator are limited to a specific age group • Comparable across populations
- 12. Age-Specific Mortality Rate Numberof deaths among persons aged 0-14 in a given year Aged 0 –14 = _________________________ 100000 years Total number of persons aged 0-14 in the same year
- 13. Age-Specific Incidence Rate Numberof new cases among persons aged 0-14 in a given year Aged 0 –14 = _________________________ 100000 years Total number of persons aged 0-14 in the same year
- 14. Age specific TBincidence rates / 1000 person / year, among migrant and native populations 0.00 1.00 2.00 3.00 4.00 0-4 5-14 15-44 45-64 65+ TBIncidenceRate Age group Native population TB Incidence rate / 1000 Imigran TB Incidence rate / 1000
- 15. Adjusting (Standardizing) Rates Allows comparisons of rates between populations that differ by variables that can influence the rate (e.g., age) Two methods: Direct method Indirect method
- 16. Adjusted Rates Advantages Summarystatement Differences in group composition “removed” hence allow unbiased comparison Disadvantages Fictional rates Absolute magnitude dependent on standard population chosen Opposing trends in subgroups masked
- 17. Rate of Disease Isa basic measure of disease occurrence because it most clearly expresses probability or risk of disease1 in a defined population2 over a period of time3 Incidence rate Mortality rate
- 18. Crude TB incidencerates in native and immigrant populations Male pop. (thousands) TB Cases TB Incidence rate / 1000 Male pop. (thousands) TB Cases TB Incidence rate / 1000 Total 21300 19380 0.91 175 148 0.84 Native poulation Immigrant
- 19. Age specific TBincidence rates in native and immigrant population Male pop. (thousands) TB Cases TB Incidence rate / 1000 Male pop. (thousands) TB Cases TB Incidence rate / 1000 0-4 1917 1150 0.60 28 30 1.07 5-14 2556 186 0.07 33 6 0.18 15-44 6816 1686 0.25 91 44 0.48 45-64 6177 7358 1.19 36 14 0.38 65+ 3834 9000 2.35 32 9 0.28 Total 21300 19380 0.91 220 103 0.47 Age Native poulation Immigrant
- 20. Age specific TBincidence rates / 1000 person / year, among migrant and native populations 0.00 1.00 2.00 3.00 4.00 0-4 5-14 15-44 45-64 65+ Age group TBIncidenceRate Native population TB Incidence rate / 1000 Imigran TB Incidence rate / 1000
- 21. Why is thediscrepancy? Native population relative age structure 8% 19% Immigrant population relative age structureAge group (y) 0-49% 5-14 16% 5% 100% 12% 100% 52% Total 45-64 15-4432% 29% 18% 65+
- 22. Problems with CrudeRates ♦ Crude rates may obscure the fact that subgroups of the population exhibit significant differences in risk ♦ Before making any conclusion, we need to compare age specific rates for these two populations; all age specific rates for Immigrants are higher than those for Natives. ♦ Since rates change dramatically with age, differences in age distributions between populations need to be accounted for before attempting to compare their risks
- 23. (Population: native or immigrant) Confounding Tbincidence (Age structure) Observed association True associationUnobserved association
- 24. Standardization of Rates •Also referred to as adjusting rates •Used to reduce distortion in comparisons between crude rates •Permits comparison of event occurrence in two or more study groups which are adjusted for differences in the variable of interest of the groups
- 25. So why standardiserates? To facilitate comparison between two or more different populations (geographical areas, different hospital populations, experimental groups That differ in composition e.g. age or sex distribution In this case crude rates may be misleading
- 26. How or why? Taking an example of CMR, risk of dying depends very much on age Age specific death rates are high for infants and very old people and low for middle age groups* So CMR and overall incidence rates will depend on age composition of the population concerned.
- 27. Direct standardization Appliedwhen variable-specific rates from each of the populations under study are applied to the standard population The outcome comprise the variable-adjusted rates Examples: age-sex adjusted mortality, morbidity or fertility rates
- 28. Direct Adjustments ofRates Requires a standard population, to which the study population’s variable-specific (e.g. age-specific) rates can be applied Choice of the standard population may affect the magnitude of the age-adjusted rates, but not the ranking of the populations
- 29.
- 30. 1. Define standardpopulation both combined, bigger of the two, national or international
- 31. Community A CommunityB STANDARD (both combined) Age (year) Population Population Population Under 1 1,000 5,000 1,000 + 5,000 = 6,000 1 – 14 3,000 20,000 15 – 34 6,000 35,000 35 – 54 13,000 17,000 55 – 64 7,000 8,000 Over 64 20,000 15,000 All ages 50,000 100,000 Defining the standard population
- 32. 2. Obtain thenumber of cases or deaths we would expect in the standard population Apply the age (or other category) specific rates of the index population to the numbers in each age group of the standard population Obtain the number of cases or deaths we would expect in the standard population if the index rates applied to the standard population
- 33. Community A CommunityB Age (year) Population Deaths Death Rate (per 1000) Population Deaths Death Rate (per 1000) Under 1 1,000 15 15/1000 x 1000 =15.0 5,000 100 100/5000 x 1000 = 20.0 1 – 14 3,000 3 3/3000 x 1000=1.0 20,000 35 15 – 34 6,000 6 35,000 35 35 – 54 13,000 52 17,000 85 55 – 64 7,000 105 8,000 160 Over 64 20,000 1,600 15,000 1,350 All ages 50,000 1,781 35.6 100,000 1,740 17.4 Calculating age-specific death rates of each study population
- 34. Expected number ofcases or deaths in the standard population Age (years) Standard population Death rate in A (per 1,000) Expected deaths at A’s rate Death rate in B (per 1,000) Expected deaths at B’s rate Under 1 6,000 15.0 15/1000 x 6,000 = 90 20.0 20/1000 x 6,000 = 120.0 1 – 14 23,000 1.0 0.5 15 – 34 41,000 1.0 1.0 35 – 54 30,000 4.0 5.0 55 – 64 15,000 15.0 20.0 Over 64 35,000 80.0 90.0 Total 150,000 35,6 17.4 Age – adjusted death rate (per 1000)
- 35. 3. Obtain thetotal number of expected cases or deaths Add the expected cases or deaths over the age groups The total number of expected cases or deaths is obtained
- 36. Total number ofexpected cases or deaths Age (years) Standard population Death rate in A (per 1,000) Expected deaths at A’s rate Death rate in B (per 1,000) Expected deaths at B’s rate Under 1 6,000 15.0 90 20.0 120.0 1 – 14 23,000 1.0 23 0.5 11.5 15 – 34 41,000 1.0 41 1.0 41.0 35 – 54 30,000 4.0 120 5.0 150.0 55 – 64 15,000 15.0 225 20.0 300.0 Over 64 35,000 80.0 2,800 90.0 3,150 Total 150,000 35,6 3,299 17.4 3,772.5* Age – adjusted death rate (per 1000)
- 37. 4. Obtain variable-standardizedrate Divide the total expected cases or deaths by the total standard population A standardised rate is obtained. In our example, an age-adjusted death rate is obtained
- 38. Age-adjusted death rates Age (years) Standard population Death rate inA (per 1,000) Expected deaths at A’s rate Death rate in B (per 1,000) Expected deaths at B’s rate Under 1 6,000 15.0 90 20.0 120.0 1 – 14 23,000 1.0 23 0.5 11.5 15 – 34 41,000 1.0 41 1.0 41.0 35 – 54 30,000 4.0 120 5.0 150.0 55 – 64 15,000 15.0 225 20.0 300.0 Over 64 35,000 80.0 2,800 90.0 3,150 Total 150,000 35,6 3,299 17.4 3,772.5 Age – adjusted death rate (per 1000) 3299/150000 x 1000 = 22.0 3772.5/150000 x 1000 = 25.0
- 39. Community A CommunityB Age (year) Population Deaths Death Rate (per 1000) Population Deaths Death Rate (per 1000) Under 1 1,000 15 15.0 5,000 100 20.0 1 – 14 3,000 3 1.0 20,000 35 1.0 15 – 34 6,000 6 1.0 35,000 35 1.0 35 – 54 13,000 52 4.0 17,000 85 5.0 55 – 64 7,000 105 15.0 8,000 160 20.0 Over 64 20,000 1,600 80.0 15,000 1,350 90.0 All ages 50,000 1,781 35.6 100,000 1,740 17.4 Unadjusted rates
- 40. In studiesof morbidity standardized rate my be: prevalence or incidence rate In studies of mortality it will be standardized mortality rate.
- 41. SUMMARY: Direct standardization Requiresstratum specific (e.g. age-specific) rates in the index populations The number of cases observed in the study population should be large enough to give meaningful stratum-specific rates necessary for direct standardization
- 42. Indirect standardization ofrates Variable-specific rates from the standard population are treated to the study populations to give the standardized rates Involves the use of known specific rates applied to the actual (observed) population characteristic of interest being compared to generate expected events
- 43. Indirect standardization isused when Age-specific death rates for the index population are unknown The index populations or number of deaths are too small for calculating stable age-specific rates It has been called ‘the mirror image of the direct method’
- 44. Standardized incidence ratio Obtained when observed events are divided by the expected events In case of death it will be the standardized mortality ratio (SMR) SMR = Observed events in the index population Expected events from the standard rates SMR = Total observed deaths in a population Expected deaths in a population
- 45.
- 46. Define the setof standard variable-specific rates Example: a set of standard age specific rates
- 47. 1. Apply thestandard age specific rates to the population in the corresponding age group of the index population to get the expected number of cases or deaths in each group 2. Add the expected cases or deaths over the age groups 3. Divide the total observed cases or deaths by the total expected to get (SMR) 4. The crude rate in the standard population multiplied by the SMR gives the standardized rate in the index population.
- 48. Population of CommunityA by Age and Standard Death Rates Age (years) Population in A Standard death rate (per 1,000) Under 1 1,000 20.0 1 – 14 3,000 0.5 15 – 34 6,000 1.0 35 – 54 13,000 5.0 55 – 64 7,000 20.0 Over 64 20,000 90.0 Total 50,000 17.4 Assume that the age-specific death rates in population B are the standard
- 49. Population and ExpectedDeaths of Community A by Age Age (years) Population in A Standard death rate (per 1,000) Expected deaths in A at standard rates Under 1 1,000 20.0 20.0 1 – 14 3,000 0.5 1.5 15 – 34 6,000 1.0 6.0 35 – 54 13,000 5.0 65.0 55 – 64 7,000 20.0 140.0 Over 64 20,000 90.0 1,800.0 Total 50,000 17.4 2,032.5 SMRA = 1781 / 2032.5 = 0.876 SMRB = 1.0
- 50. Standardized Mortality Ratio Ifthe SMR is greater than 1, more deaths have occurred than anticipated If the SMR is less than 1, fewer deaths have occurred than anticipated
Editor's Notes
- #4 A proportion is a special ratio in which the numerator is part of the denominator E.g. Proportion of girls in the first year PG class at MUHAS = No. of girls in first year PG class Total No. of students in the first year PG class
- #7 Crude mortality rate Not a true rate (actually a ratio). Numerator not part of the denominator
- #8 E.g. CBR = No. of livebirths in a year x 1000 Total population The rate is crude because it does not take into account the risk of giving birth according to age and sex differences
- #25 ‘Standard’ - a model Standardization = the process of complying (or evaluate by comparing) with a standard