Treatment completion rates for ltbi, arkansas, 2006-2010: What do we do with negative findings ?

TREATMENT COMPLETION RATES FOR LTBI, ARKANSAS, 2006-2010: WHAT DO WE DO WITH NEGATIVE

FINDINGS ?

Mukasa LN, MBChB PhD., Bates JH, MD., Phillips J. MD., Abernathy RS, MD.,

Karpoff E. BSN RNP., Diatta N., Patil N, MD MHSA.

Arkansas Department of Health

Background/Statement of Problem Treatment completion for LTBI was

original focus area for program evaluation in Arkansas, 2006

LTBI is reportable in Arkansas Over the years, treatment LTBI remained

priority Thirty seven percent of the cases, 2004-

2007 were missed opportunities We sought to increase LTBI Tx

completion to 65% by 2010

Background…

Average 2500 LTBI cases diagnosed per year

Reporting LTBI initially incomplete using paper forms

In 2005, we predicted that the TB case rate would be

2.0 per 100, 000 population in 2010 if interventions at time were maintained

We sought to accelerate TB elimination by aggressively treating LTBI

A Missed TB Opportunity was defined as a TB case:

Known to have a prior LTBI diagnosis Known to be HIV positive before TB

diagnosis From established targeted intervention

groups Nursing Homes Prisons Health Care Workers Foreign-born students

Age less than 5 years Reported After Death

Prediction of Arkansas TB Case Rates

2006-2010

Year TB Case Rate

Predicted Observed

2006 47 36

2007 32 37

2008 28 29

2009 24 28

2010 20 27

*Rate Per million population

Aims/Objectives

1) Evaluate Completion LTBI Treatment 2004-2010

2) Illustrate concept second generation surveillance Case rate vs. Treatment LTBI Clustering vs. Treatment of LTBI

3) Discuss next steps

Methods

Stakeholders included clinicians, colleges, poultry workers, correction institutions and nursing homes

LTBI cases came from targeted testing and contact investigation

Cases LTBI were captured using a standard form

Analysis was approached in 3 ways: Treatment completion rates for LTBI Construction of a missed opportunity

measure Genotype cluster analysis

Data Flow

Web-EnabledTB Registry

TB Contact

sLTBI Cases

E-forms E-forms

Evaluation Treatment LTBI

• Contact Investigations

• Targeted Testing

Arkansas 75 Counties Within 5 Regions:A Centralized Public Health Infrastructure

Little Rock

Predictive Modelof LTBI Treatment to include Regions is Pending

Results

Treatment completion rates for LTBI ranged from 50 to 55% during the 5-year study period

35-40% of all TB cases were missed opportunities

13.5% of all cases were known prior LTBI

64% of case-isolates were in genotype clusters

TB Cases Reported from Prior LTBI, Arkansas, 2004-2010

Year Total Cases

Prior LTBINumber Percent

2010 78 7 9.0

2009 82 11 13.4

2008 83 21 25.3

2004-2007 454 55 12.1

Overall 697 94 13.5

TB Trends by Treatment Completion for LTBI, Arkansas, 2004-2010

Year TB Cases

TB Case Rate**

Foreign-born (%)

Completion TLTBI (%)

2010 78 2.7 30.8 48.0*

2009 82 2.8 28.0 55.7

2008 83 2.9 22.9 54.6

2007 106 3.7 29.2 53.2

2006 102 3.6 31.4 50.5

2005 114 4.1 29.0 52.8

2004 132 4.8 36.4 49.3

2003 127 4.7 18.9

2002 136 5.0 14.7

2001 162 6.1 17.9

2000 199 7.8 14.6TLTBI =Treatment LTBI * Provisional Data** Cases per 100,000 population

Genotype Clustering, Arkansas, 2005-2010, (n=401)

64%36%

Spoligotype-MIRU method

ElectronicContact Form

2-Page Electronic LTBI Form

The TB Trends in Arkansas and the USare in similar direction

Are we constructing another U-shaped Curve of Concern ?

Conclusions/Recommendations

The goal of increasing LTBI treatment completion rates from 50% to 65% was not achieved. Complete data analysis is pending

We could not link the modest decline in TB incidence, 2006-2010 to treatment LTBI

The level of clustering by genotype suggests substantial ongoing TB transmission

Deployment of two electronic forms for TB contacts and LTBI will greatly improve surveillance and real-time analysis of LTBI in Arkansas The introduction of the long awaited “game changers”, namely IGRA and 3-month regimen for LTBI is ongoing in Arkansas.

The new tools in deployment, we believe, will reverse the negative findings in our our study

Next Steps

Computerize all LTBI records 2006-2010

Construct predictive models Treatment Completion

Logistic RegressionTreatment completion (Y/N)=region + gender+ age + US-born status, etc…

Cox’s Proportional Hazards Model Curves Months on Isoniazid by group e.g region, US-born status, etc..

Study and adopt interventions from other programs