Interactive table: See how SA’s provinces compare on the 90-90-90 targets

New estimates from the Thembisa model (a leading mathematical model of HIV in South Africa) suggest that while KwaZulu-Natal is doing well in its HIV response, the North West province is struggling.

90-90-90

One widely endorsed set of measures to track a country or province’s HIV response is UNAIDS’s so-called 90-90-90 targets – a set of targets also adopted in South Africa’s National Strategic Plan for HIV, TB and STIs 2017 – 2022.

Related Posts

• From the frontlines: 20 years of fighting HIV in Khayelitsha
• HIV and circumcision: Where are we at the end of 2020?
• HIV death rate in SA men much higher than in women
• HIV in SA: Five graphs that tell the story
• HIV in SA: Seven graphs that tell the story
• HIV in Umkhanyakude: Impressive numbers, but living with HIV difficult amid socio-economic hardship
• In graphics: Will South Africa meet UNAIDS’s HIV targets for 2020?
• In-depth: Getting Gauteng’s HIV and TB response back on track
• New HIV/TB plan delayed by a year because of COVID-19
• Nuances in SA’s HIV Epidemic: Seven Graphs that Tell the Story
• Opinion: When civil society can’t do its work, adherence to HIV treatment dips.
• SA compares well with US on some HIV metrics
• U=U: We should put people living with HIV at the centre of HIV prevention efforts
• WORLD AIDS DAY: How stigma is impacting the 90-90-90 goals

The targets are that by 2020:

1. 90% of people living with HIV must be diagnosed,
2. 90% of those diagnosed must be receiving sustained antiretroviral therapy, and
3. 90% of those diagnosed must achieve viral suppression (i.e. there must be so little virus in their blood that standard tests do not pick it up).

How are SA’s provinces doing?

Province	1st 90	2nd 90	3rd 90	ART coverage	% of all HIV+ virally suppressed
EC	90.4475	64.9638	89.092	58.7599	52.3496
FS	89.4938	72.6132	93.4507	64.986	60.7294
GP	89.0756	66.2594	87.624	59.021	51.717
KZN	93.2086	75.7269	94.059	70.5842	66.3911
LM	90.9255	67.6311	87.4936	61.4954	53.807
MP	90.4651	70.6798	91.1466	63.9417	58.2814
NC	90.6072	68.2079	89.8192	61.8036	55.5125
NW	89.5832	60.5223	90.4545	54.2195	49.0453
WC	91.0629	66.6988	93.4046	60.7381	56.732

You can click the column headings to sort by the related indicator.

In addition to the 90-90-90 targets we added two extra columns to the right:

• The ART coverage column shows the percentage of all people living with HIV (diagnosed and undiagnosed) who are on treatment – unlike the second 90 which reflects just the percentage of those who are diagnosed who are on treatment.
• The column on the far right shows the percentage of all people living with HIV who are virally suppressed – unlike the third 90 which reflects only the percentage of people both diagnosed and on treatment who are virally suppressed. Of the five, this last column is probably the single column that tells us the most about the state of a province’s HIV response (technically it is the product of the three 90s).

Key takeaways

• On the 90-90-90 measures KwaZulu-Natal performs better than any of South Africa’s other provinces. This is particularly impressive given that the province has a large population and much of the province is rural. The Free State does second best on this set of indicators.
• The worst performing of the nine provinces, at least based on these indicators, is the North West. This is not surprising given well-documented governance issues in the province’s health department. That Gauteng ranks second worse may come as a surprise to some – especially when contrasted with the more rural KwaZulu-Natal. The Eastern Cape also clearly still faces very significant challenges – although that is probably less surprising.
• Across the country HIV testing efforts have been impressive with all provinces exceeding or very close to meeting the first of the 90-90-90 targets.
• The one clear short-coming across provinces is that many people who test positive either stop taking treatment or never start (the second 90). Helping more people who are living with HIV to start and stay on treatment must thus be a top priority for provincial departments of health.

Note: The numbers quoted in the table are the 2019 mean point estimates from the Thembisa model version 4.3 (confidence intervals are provided in the published outputs on the Thembisa website). Thembisa also includes estimates for 2020, but these are substantially more uncertain than the 2019 estimates. For viral load suppression we used the 1 000 threshold rather than 400 (Thembisa provides both).

Like what you are reading? CLICK HERE to sign up for our newsletter.