Step 1. Select the Focus of Your Query

• You can choose any of the following foci:

  • Total number of patients
  • Inpatient admissions
  • Outpatient visits
  • Inpatient costs
  • Outpatient costs
  • Total costs

The first option (total number of patients) is useful if you want to see the distribution of demographic or clinical characteristics in the database. You can choose multiple options; this will result in multiple tables.

The inpatient cost analyses are based on an estimate of a "unit cost" per inpatient day of $1,760. The outpatient cost analyses are based on an estimate of a unit cost per visit of $189. These figures are based on estimates calculated in the HIV Cost and Service Utilization Study, and adjusted for inflation. Inpatient cost estimates per person were calculated by multiplying the total number of inpatient days by the cost per day figure. Persons with no inpatient utilization were assigned zero inpatient costs. Outpatient cost estimates per person were calculated by multiplying the total number of outpatient visits by the outpatient cost per visit amount. Total costs are the sum of inpatient and outpatient costs. Costs of prescription medications are not included in this estimate. Because a single unit cost is used, the resulting cost estimates do not incorporate any adjustment for differences in the amount or intensity of services delivered.

Step 2. Choose a Statistic

• For inpatient admissions and outpatient visits, you can choose the mean, the median, the percent with no use, or a contingency table. For inpatient, outpatient, and total costs, you may choose the mean or the median. The mean is the arithmetic average. The median is the midpoint - half the cases fall above this point and half below. The median is provided because the mean can be strongly influenced by extreme values. The mean and median provide useful summary statistics. The percent with no use provides information on who is not using a type of service.
  
  Finally, the contingency table option will categorize the focus variable (i.e., inpatient, outpatient, or ER) and cross-tabulate this with other variables that you select (one variable at a time). This option provides the most detailed information. It will present the number of patients in each combination of categories of focus variable and other variable. For example, a table of inpatient admissions by gender will show numbers of men and numbers of women in each category of inpatient admissions.
  
  

You can choose any combination of these options. The mean, median, and percent with no use will be presented in one table, and the contingency table will be shown in a second, separate table.

Once you have chosen a focus variable and a statistical presentation, click on >>Next>> button to go to the next window. There, you can choose classification variables.

Step 3. Choose Classification Variables

• To look at resource use statistics for various subgroups, click the square next to each variable. You can select from the following list:
• age
• value of first CD-4 count in observation period
• value of first viral load measure in observation period
• gender
• race
• HIV risk group (e.g., men who have sex with men, intravenous drug use)
• insurance status (private, public, none)
• PI: whether or not the patient was prescribed a Protease Inhibitor
• NRTI: whether or not the patient was prescribed a Nucleoside Reverse Transcriptase Inhibitor
• NNRTI: whether or not the patient was prescribed a Non-Nucleoside Reverse Transcriptase Inhibitor
• HAART: whether or not the patient was prescribed highly active antiretroviral therapy

PI, NRTI, and NNRTI are classes of antiretroviral medications. HAART refers to the "AIDS cocktails," combinations of several antiretroviral medications. CD-4 and viral load measure immune functioning.

You can choose none, one, or several of these variables. If you choose multiple classification variables, the system will display results separately for each variable.

Clicking on the >>Next>> button will then display a table of results for your query.

HIVNet is based on aggregate statistics tables to speed up data transfer and to protect the confidentiality of individual records. Therefore, not all possible queries can be addressed. All tables generated by HIVNet are descriptive statistics without any testing for statistically significant differences. Statistics based on 10 or fewer cases are not reliable and are suppressed.