HIV/AIDS Prevalence and Mortality Technical Notes - Minnesota Dept. of Health

HIV/AIDS Prevalence and Mortality Technical Notes 2005

Surveillance of HIV/AIDS

Formatted for print: HIV/AIDS Prevalence and Mortality Technical Notes (PDF: 152KB/6 pages)

The Minnesota Department of Health (MDH) collects case reports of HIV infection and AIDS diagnoses through a passive and active HIV/AIDS surveillance system.  Passive surveillance relies on physicians and laboratories to report new cases of HIV infection or AIDS directly to the MDH in compliance with state (1).Active surveillance conducted by MDH staff involves routine visits and correspondence with select HIV clinical facilities to ensure completeness of reporting and accuracy of the data.

Factors that impact the completeness and accuracy of HIV/AIDS surveillance data include: availability and targeting of HIV testing services, test-seeking behaviors of HIV-infected individuals, compliance with case reporting, and timeliness of case reporting.   Certain events have also impacted trends in HIV/AIDS surveillance data.  For example changes over time in the surveillance case definition (most notably the 1993 expansion of the case definition for adults and adolescents(2))have resulted in artificial jumps in AIDS case counts at the time the new definition went into effect or in the preceding year because changes in case definition allowed for retrospective diagnoses.

Vital Status of HIV/AIDS Cases

Persons are assumed alive unless the MDH has knowledge of their death.  Vital status information is updated by monthly visits to select reporting facilities, correspondence with other health departments, annual death certificate reviews, and periodic matches with the National Death Index.  “AIDS deaths” refers to all deaths among AIDS cases regardless of the cause of death.

Place of Residence for HIV/AIDS Cases

Persons are assumed to be residing in Minnesota if their most recently reported state of residence was Minnesota and the MDH has not received notice of relocation outside of the state.  Likewise, a person’s county or city of residence is assumed to be the most recently reported value unless the MDH is otherwise notified.  Residence information is updated through standard case reporting, monthly visits to select reporting facilities and/or correspondence with other state health departments.  Persons diagnosed with HIV infection while imprisoned in a state correctional facility are included in the data presented unless otherwise noted (federal and private prisoners are excluded).  Residential relocation, including release from state prison, is difficult to track and therefore data presented by current residence must be interpreted in this light.  Data on residence at time of diagnosis are considered more accurate, limited only by the accuracy of self-reported residence location.

Data Tabulation and Presentation

Unless otherwise noted, data analyses exclude persons diagnosed in federal or private correctional facilities (inmates generally are not Minnesota residents before incarceration and do not stay in Minnesota upon their release), infants with unknown or negative HIV status who were born to HIV positive mothers.  Data include HIV-infected refugees who resettled in Minnesota as part of the HIV-Positive Refugee Resettlement Program.

The HIV/AIDS surveillance system is a live database that is continuously updated to reflect the most current information available. Variables such as current state of residence are over-written when updates are made.  Annual archive files were initiated in 2001.  Thus, the numbers of HIV/AIDS cases residing in Minnesota in 2000 and 2001 were estimated using the current state of residence variable while the number in previous years (1990-1999) was estimated using state of residence at time of diagnosis, vital status, and date of death variables. The number of HIV/AIDS cases alive in a certain year was calculated by summing cases with an HIV/AIDS diagnosis in that year or prior whose vital status in 2001 was “alive” or whose date of death was either after the calendar year of interest or missing.

Mode of Exposure Hierarchy

All state and city HIV/AIDS surveillance systems funded by the Centers for Disease Control and Prevention use a standardized hierarchy of mode of exposure categories. HIV and AIDS cases with more than one reported mode of exposure to HIV are classified in the exposure category listed first in the hierarchy. In this way, each case is counted as having only one mode of exposure.  The only exception to this rule is the joint risk of male-to-male sex (MSM) and intravenous drug use (IDU), which makes up a separate exposure category in the hierarchy.  The following is a list of the hierarchy for adolescent/adult HIV/AIDS cases:

  1. MSM
  2. IDU
  3. MSM/IDU
  4. Hemophilia patient
  5. Heterosexual contact
  6. Receipt of blood transfusion or tissue/organ transplant
  7. Other (e.g. needle stick in a health care setting)
  8. Risk not specified.

The following is the list of the hierarchy for pediatric HIV/AIDS cases:

  1. Hemophilia patient
  2. Mother with HIV or HIV risk
  3. Receipt of blood transfusion or tissue/organ transplant
  4. Other
  5. Risk not specified.

Heterosexual contact is only designated if a male or female can report specific heterosexual contact with a partner who has, or is at increased risk for, HIV infection (e.g. an intravenous drug user).  For females this includes heterosexual contact with a bisexual male (mainly due to the elevated prevalence of HIV infection among men who have sex with men).

“Risk not specified” refers to cases with no reported history of exposure to HIV through any of the routes listed in the hierarchy of exposure categories.  These cases include persons who have not yet been interviewed by MDH staff; persons whose exposure history is incomplete because they died, declined to be interviewed, or were lost to follow-up; and persons who were interviewed or for whom follow-up information was available but no exposure was identified/acknowledged. 

The growing number of cases with unspecified risk in recent years is, in part, artificial and due to interviews that have not yet been completed.  In time, a number of these will be assigned a mode of exposure category.  However, part of the observed increase is real.  As stated above, a person must have intimate knowledge about his/her partner to meet the criteria for heterosexual mode of exposure.  Often cases will not be certain about their partners’ HIV status or risk.  Additionally, the perception of social stigma presumably decreases the likelihood that a person will acknowledge certain risk behaviors, particularly male-to-male sex or injecting drug use.  Thus, if the true numbers of cases due to heterosexual contact, MSM, and/or IDU increase, a larger number of cases without a specified risk would be expected. 

A recent study by the Centers for Disease Control and Prevention used statistical methods to redistribute risk among female HIV/AIDS cases with unspecified risk (3).The results are helpful but are based on national data that are not necessarily applicable to the state or local level.  Speculation regarding the distribution of risk behaviors among those with unspecified risk is difficult, especially in men, for whom even a national study is not available.

Re-distribution of Mode of Exposure

In 2004 the Minnesota Department of Health began estimating mode of exposure for cases with unspecified risk in its annual summary slides. Estimation was done by using the risk distribution for living cases with known risk by race and gender and applying it to those with unspecified risk of the same race and gender. There were two exceptions to this method, African-born cases and Asian/Pacific Islander women. For both African-born and Asian/Pacific Islander women a breakdown of 95% heterosexual risk and 5% other risk was used. For African-born males a breakdown of 5% male-to-male sex, 90% heterosexual risk, and 5% other risk was used. These percentages are based on epidemiological literature and/or community experience.

Below is an example of how the process worked for white, African American and African-born females:

Living Cases Among Females in 2005
Heterosexual n (%†)
IDU n (%†)
Other n (%†)
Unspecified n
Total N
186 (73)
57 (22)
11 (4)
177 (69)
63 (25)
16 (6)
53 (85)
0 (0)
9 (15)

† Percent of those with know risk.

Female Cases with Estimated Risk
Total N
(.73*89) + 186 = 251
(.22*89) + 57 = 77
(.04*89) + 11 = 15
(.69*118) + 177 = 259
(.25*118) + 63 = 92
(.06*118) + 16 = 23
African-born ‡
(.95*248) + 53 = 289
(.05*248) + 9 = 21

‡ Used a distribution of 95% heterosexual and 5% other.

Definitions Related to Race/Ethnicity

When data are stratified by race, Black race is broken down into African-born and African American (not African-born) based on reported country of birth.

The terms “persons of color” and “non-Whites” refer to all race/ethnicity categories other than White (Black, Hispanic, American Indian, and Asian/Pacific Islander).

Interstate De-Duplication Project (IDEP)

            In 2005, the Minnesota Department of Health (MDH) participated in RIDR. RIDR is a CDC project aimed at eliminating duplicate reports of HIV and AIDS cases among states. Each case of HIV and AIDS is assigned to the state (or states when the diagnosis of HIV and AIDS occurs in two different states) where a person was first diagnosed. RIDR was the second such de-duplication initiative by CDC. The first initiative, IDEP, looked at cases reported through December 31, 2001. RIDR included cases from January 1, 2002 through June 30, 2005. Through this project, MDH identified 50 cases of HIV infection (including AIDS at first report) and 12 AIDS cases whose first diagnosis was not in Minnesota. These cases were previously considered as diagnosed in Minnesota and were counted in the cumulative number of cases diagnosed in Minnesota. As such, the change of “ownership” (where the case was diagnosed) has reduced both cumulative and yearly totals for Minnesota. Additionally, MDH also identified 142 cases that no longer live in Minnesota.

(1) Minnesota Rule 4605.7040 (return to text)
(2) MMWR 1992;41[no.RR-17]:1-19 (return to text)
(3) MMWR 2001; 50(RR-6):31-40 (return to text)

Updated Friday, September 16, 2016 at 11:22AM