Thursday, August 27, 2009



SF DPH: 2008 HIV Stats
Fell 45% From 2004

The latest annual HIV epidemiology report for the USA's AIDS Model City came out last week, and the central marker of the current state of HIV infections in the city, new HIV infections, fell to 434 in 2008. However, there are many caveats to that stat, as the report makes clear, on page 17:

The number of HIV/AIDS cases diagnosed was relatively stable between 2006 and 2007 and declined in 2008 (Figure 1.2). In general, there is a delay between the time a person is diagnosed with HIV disease and the time that person is reported to the health department. For this reason, numbers of cases diagnosed in recent years were often lower because of persons diagnosed in these years who have not yet been reported.

There is a statistical method developed by the Centers for Disease Control and Prevention to take reporting delay into account when examining the trends in HIV disease diagnosis. If we apply this statistical method to our data to adjust for reporting delay, the number of persons diagnosed with HIV between 2006 and 2008 would be steadily increasing. Therefore, careful attention should be paid to the actual numbers of new diagnoses over time.

My interpretation of the two paragraphs, as a longtime observer of SF's HIV numbers, is that the health department has a few ways of not coming right out and stating that the numbers continue to be flat or declining. For the past 3-4 years, even the SF Chronicle has duly reported on the steady drops in new infections, but the DPH will go to great lengths to find the worst and gloomiest HIV data. DPH is incapable of actively recognizing the good news about falling infections here.

In addition, trends in HIV diagnosis using case reporting data should be interpreted with caution. These data include HIV infected persons who have been tested or who are in care and do not include persons who are not aware of their infection. These data also do not necessarily reflect trends in new HIV infections.

When the stats in 1999/2000 were supposedly increasing at sub-Saharan levels, DPH did not caution that the numbers then should be looked at with caution or that the stats were not a true indications of new infections.

Our estimate of HIV incidence in San Francisco (see Executive Summary) indicates that the number of new HIV infections remain relatively stable between 2006 and 2007. Therefore, if the upward trends in HIV diagnosis between 2006 and 2008 estimated after accounting for reporting delay is true (which can only be validated in a few years by examining the data retrospectively), it may reflect increases in HIV testing and diagnosis among those who were previously unaware of their infection.

Could the DPH be more careful to tread lightly around the fact that for years most of the increases in new infections are due to increased testing and surveillance?

So what are the actual numbers for new infections? Here they are, on page 18:

Table 1.3 shows the characteristics of persons diagnosed with HIV between 2004 and 2008. The majority were male, white, aged 25-49 years, and MSM. There was a slight increase in the proportion of persons of color.

Table 1.3 Characteristics of persons newly diagnosed with HIV between 2004 and 2008
San Francisco: 2004 / 2005 / 2006 / 2007 / 2008
Total Number: 798 / 642 / 523 / 518 / 434

A drop of 45%. Even accounting for reporting delays and such, I see a dramatic drop over this five-year period. Just don't expert the AIDS authorities to say so.

I have previously stated my belief that new HIV infections for San Francisco have long been on the decline, primarily due to reduced infectiousness among positive persons on the anti-retroviral therapies, ART. With so many pozzies on cocktails, and adhering to their regime of proper dosing and regular viral load and STD monitoring, there's little question about the integral role cocktails play in preventing new infections.

The 2008 epi report shares this stat, on page 40, reinforcing the notion cocktails equal fewer infections:

Overall, 88%-92% of persons living with AIDS received ART.

That's a lot of poz people on drug cocktails and, in my opinion, a key reason why the HIV rate in SF is low, and keeps declining. Click here to read the full 2008 report.

1 comment:

Anonymous said...

But what about the new confidential requirement vs. the old anonymous standard?

I know that after testing went confidential at most testing centers I was hesitant to get tested for over a 18 months, while I normally get tested every six.

Is it possible that the new confidentiality requirement is resulting in suppression of people seeking tests?

Does the report have comparative information on the number of tests given between 2004 and 2008?