27.01.2015 Views

Guidelines for second generation HIV surveillance - World Health ...

Guidelines for second generation HIV surveillance - World Health ...

Guidelines for second generation HIV surveillance - World Health ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Appendix A: Interpret prevalence trends<br />

across multiple sites<br />

You will commonly find two problems related to analysing data from multiple sites when interpreting data<br />

on biological or behavioural trends:<br />

• improper aggregation – combining small samples improperly<br />

• inconsistent survey sites.<br />

Improper aggregation<br />

When looking at trend data, it is tempting to pool data from multiple sites to obtain a larger sample size.<br />

The larger sample sizes are convenient because you are trying to detect statistically significant trends over<br />

time, but this is not the best solution.<br />

What are the issues<br />

The disadvantage of pooling or aggregating data across sites is that the differences between the sites are<br />

obscured. Your result may not be meaningful and may even be misleading.<br />

What is gained in sample size by aggregating data may be lost in the inability to see potentially important<br />

local patterns. Figure A.1 illustrates how the conclusions drawn from a combined trend line <strong>for</strong> all sites is<br />

different from those of the individual sites.<br />

• The combined-sites trend line misses the rising prevalence in site A and the overall higher prevalence<br />

in site B.<br />

• It also misses the decline in prevalence in site D.<br />

• Some of the fluctuations in the individual trend lines, such as at site F, are likely to be the result of small<br />

sample sizes. Those effects are smoothed out when the data are aggregated.<br />

Figure A.1. Comparing the trend lines of all sites pooled versus individual sites<br />

0.07<br />

<strong>HIV</strong> prevalence<br />

0.06<br />

0.05<br />

0.04<br />

0.03<br />

0.02<br />

Site A<br />

Site B<br />

Site C<br />

Site D<br />

Site E<br />

Site F<br />

All Sites<br />

0.01<br />

0.0<br />

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007<br />

54

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!