Department of Biostatistics

People

Prospective Students

Academics

Research

News & Events

Calendar

Consulting

Employment Opportunities

Resource Quicklinks

Computing Environment

Contact

Description:

This program is used to project the incidence and prevalence of a chronic disease in the United States, such as Alzheimer's disease, which affects the elderly. The potential impact of public health interventions that reduce disease incidence rates and/or disease death rates also can be evaluated. Projections are made through the year 2050.

Calculations are performed using age- and gender-specific U.S. mortality rates and U.S. Census Bureau projections of future population sizes. The U.S. mortality rates were assembled from national vital statistics records and the U.S. population forecasts were assembled from the U.S. Census Bureau's "middle series" estimates. In addition to mortality rates and population projections, the program requires user-specified parameters, including age-specific incidence rates. These user-specified parameters are described below. Further details of the methodology can be found in the following two references.

References:

1. Brookmeyer R, Gray S and Kawas C: Projections of Alzheimer's disease in the United States and the public health impact of delaying disease onset. American Journal of Public Health 88(9), 1337-1342, 1998.

2. Brookmeyer R, Gray S: Methods for projecting the incidence and prevalence of chronic diseases in ageing populations: application to Alzheimer's disease. Statistics in Medicine 19(11-12):1481-1493, 2000.

Instructions:

To use this program, you first need to download the following files (typically right click on link and choose "save as"):

1. projec.f: Fortran 77 code used to perform the calculations.
2. dthm.dat: Data file containing U.S. age-specific mortality rates/100,000 for males ages 60-118 in calendar years 1937-2050. Rows refer to single years of age beginning with age 60 and columns refer to single calendar years beginning with year 1937. For example, in 1937 the estimated U.S. mortality rate for men age 60 was 2,531/100,000 and for men age 61 was 2,780/100,000.
3. dthf.dat: Data file containing U.S. age-specific mortality rates/100,000 for females ages 60-118 in calendar years 1937-2050. Rows refer to single years of age beginning with age 60 and columns refer to single calendar years beginning with year 1937.
4. midpopm.dat: Data file containing projected U.S. population sizes for males by single year of age 59-118 and calendar years 1995-2050. Rows refer to single years of age beginning with age 59 and columns refer to single calendar years beginning with year 1995. For example, in 1995 the estimated number of U.S. men age 59 was 1,023,480 and men age 60 was 996,879.
5. midpopf.dat: Data file containing projected U.S. population sizes for females by single year of age 59-118 and calendar years 1995-2050. Rows refer to single years of age beginning with age 59 and columns refer to single calendar years beginning with year 1995.
6. adincid.dat(optional): Example of a data file containing age-specific incidence rates/100 PY for the disease of interest for ages 60-118. (This particular file contains incidence rates that might be applied for Alzheimer's disease.) Rows refer to single years of age beginning with age 60. For example, in this data file the incidence rate for age 60 is given as 0.000873 and for age 61 is given as 0.000999.

After downloading the above files, you will need to compile the Fortran 77 code projec.f. To run the program, run the resulting executable file. You will be prompted to enter the following (some items separately for men and women):

1. Names of the files containing incidence data. The filenames should be placed in quotes and the file itself should contain one column (of length 59) with incidence rates/100 PY for the ages 60 to 118. The program automatically assumes that incidence before age 60 is 0. An example file containing incidence rates for Alzheimer's disease is adincid.dat (see 6 above).
2. The relative risk of death for diseased versus not diseased. For example, a relative risk of 1.5 implies that individuals with disease have mortality rates 50% higher than individuals of the same age without disease. This must be a positive number.
3. The calendar year in which the postulated intervention begins. This must be after 1900.
4. The relative risk of disease for treated versus untreated. For example, a relative risk of 0.90 implies the intervention reduces age-specific incidence rates by 10%. This must be a positive number.
5. The relative risk of death among those with disease for treated versus untreated. For example, a relative risk of 0.80 implies the intervention reduces disease mortality by 20%. This must be a positive number.
6. The duration of illness for which you would like prevalence estimates printed. For example, if you want prevalence estimates of all people with disease, enter 0. If you want prevalence estimates of people with disease at least 2 years, enter 2. This must be a non-negative integer.
7. If you would like age-specific prevalence rates (/100 PY) printed for a specific calendar year. If yes, you must enter the calendar year. This must be between 1998 and 2050.

Estimated incidence and prevalence projections by gender through the year 2050 will be written to the output file projec.out. In addition, if requested, estimated age-specific prevalence rates by gender for a specified year will be printed to the same output file.

Questions/Comments? E-mail smg@gene.com