The Costs of Diagnosing and Treating Tuberculosis in South Africa - - PowerPoint PPT Presentation

The Costs of Diagnosing and Treating Tuberculosis in South Africa

Sydney Rosen Center for Global Health and Development Boston University, Boston, MA Health Economics and Epidemiology Research Office (HE2RO) Wits University, Johannesburg April 17, 2013

Last Week’s Headline

• Setting the stage: Tuberculosis in

South Africa

• The cost of treating multidrug

resistant TB

• Scaling up Xpert MTB/RIF for TB

diagnosis

• A TB economics research agenda

Overview

HE RO

2

Tuberculosis in South Africa

• Leading cause of death (11.6% of all deaths 2010, v. 2.3% globally)
• Very high incidence of drug susceptible TB

– 1% or 490,000 new TB cases per year – Third highest number of new cases globally after India and China

Sources: http://www.unaids.org/en/regionscountries/countries/southafrica/; WHO, Global Tuberculosis Control 2011; WHO, South Africa TB country profile 2012; Statistics South Africa, Mortality and causes of death in South Africa, 2010; Lozano R et

• al. 2012; WHO, M/XDR-TB: 2010 Global Report on Surveillance and Response;

http://www.timeslive.co.za/local/2012/02/02/tb-is-top-killer-of-sa-blacks

• High rates TB/HIV co-infection

– Adult HIV prevalence 17.3% – ≥ 60% of TB patients co-infected with HIV – TB is the leading cause of death for HIV- infected people

• High-burden drug resistant TB country

– >10,000 confirmed MDR-TB cases in 2011 – Highest number of XDR-TB cases globally (573 cases reported in 2008)

South Africa’s TB Epidemic in Context

Indicator (2011) South Africa Russian Federation Brazil United States

Population 51 million 143 million 192 million 312 million GNI/capita (PPP) $10,710 $20,560 $11,420 $48,820 Adult HIV prevalence 17.3% 1.1% 0.3% 0.6% TB incidence/100,000 993 97 83 4 TB cases/year 389,974 159,479 84,137 10,521 Confirmed MDR-TB cases/year 10,085 13,785 556 119

Sources: http://www.unaids.org/en/regionscountries/countries/; WHO, Global Tuberculosis Control 2012; United Nations, Estimates of mid-year population: 2002 – 2011; World Bank, World Development Indicators, 2012.

TB Diagnosis

• Until now, primarily by sputum smear and culture

— Xpert MTB/RIF at all centralized laboratories by 2014—more on that soon

• 53% of pulmonary TB cases are smear-negative or unknown and

require culture

— Smear results 2+ days — Culture results 6 weeks

• Case detection rate estimated at 69%
• Resistant TB diagnosis by drug-

sensitivity testing

Source: WHO, South Africa TB Country Profile 2012

— After first line treatment failure or previous TB — Takes 42-56 days for results (Loveday et

al 2012)

TB Treatment

• Treatment of drug-sensitive TB

– Treatment lasts 6-8 months – Outpatient care by nurses at primary health clinics with treatment supervisor/direct observation – Treatment success rate 60-75%, varying with smear status, etc.

• Treatment of MDR-TB

– Treatment lasts 24 months – Guidelines call for ≤ 6 months inpatient care by doctors at specialized hospitals and remainder outpatient – Treatment success rate ≈ 50%; most of the rest die during treatment or default

• r fail treatment (Brust et al 2010)

Source: WHO, South Africa TB Country Profile 2012

The Cost of Inpatient Treatment for MultiDrug Resistant Tuberculosis

Rationale and Objectives

• No empirical estimates of the cost of treating MDR-TB in South Africa

– Only 4 estimates globally, none in Africa; range $2,791-$16,881/case

(Fitzpatrick et al 2012)

– 55% of South Africa’s TB control budget reportedly spent on MDR-TB

(WHO, South Africa tuberculosis finance profile)

– Most high burden countries rely on hospitalization, but outpatient care is recommended by WHO – Difficult to understand or improve treatment delivery or outcomes without better information about costs

• Objectives

– Estimate the cost per patient of the

inpatient phase of MDR-TB treatment – Generate comparison data for evaluating the cost-effectiveness of alternatives to the current inpatient model of care

www.phi.org

Approach

• Retrospective cohort study
• Provincial MDR-TB referral hospital in North West Province
• Obtained resource utilization and outcome data from medical records
• Data collected up to 12 months from admission or until the earliest of discharge,

abscondment, or death

• Costs estimated included hospital stay/day, drugs, lab tests, radiography, surgery

Study Sample

Recorded in hospital register Mar 2009-Feb 2010 277 Not eligible (other criteria) 12 Enrolled in study 133 Smear-status known 128 Complete resource utilization records 121 Initiated or completed MDR-TB treatment at another site 49 Non-MDR TB diagnosis 83

• Enrolled all admitted

patients with confirmed MDR-TB

• March 2009-February 2010
• Excluded transfers in or out
• N=128 (121 for costs)
• Median age 39; 45%

female; 64% unemployed

• 50% smear-, 50% smear+

at admission

• 83% previous TB
• 64% HIV infected

Treatment Outcomes at 12 Months

86% 3% 8% 3%

46% 10% 23% 21%

Cured or completed Failed Died Defaulted

Source: Farley et al 2011

Culture converted, discharged Absconded Died in hospital S ll admi ed at 12 months

Resource Utilization and Costs

Cost All patients (n=121) Smear positive (n=55) Smear negative (n=61) Number of months in hospital [SD] 3.5 [1.7] 4.2 [1.9] 3.2 [1.2] Average cost/patient $17,164 $20,440 $15,450

$16.325; 95% $380; 2% $236; 2% $223; 1% Hospital stay (personnel, infrastructure) MDR-TB drugs TB laboratory monitoring Other costs

Limitations

• Single site in one province, small sample
• Cost of MDR-TB inpatient phase only, omits 18-21 months
• utpatient care

– Drugs and laboratory monitoring will comprise a larger share of total costs if full continuation phase of treatment included

• Results conditional on being admitted at all

– Hospital admission took a mean of 111 days after sample collected for DST – Study in another province found that 40% of MDR-TB patients with HIV die within 30 days of testing (Gandhi et al 2010)

Conclusions

• Average 12-month cost of inpatient treatment for MDR-TB =

$17,164

– 67 x the full cost of treating drug-sensitive TB (Pooran et al. 2013) – 25 x the annual cost of first-line ART (Long et al 2011)

• MDR-TB treatment capacity is severely limited

– In 2011, only 56% of diagnosed MDR- and XDR-TB patients started on treatment due to shortage of hospital beds in specialized wards (WHO 2012) – Patients in our sample waited ≈ 1 month even after MDR-TB diagnosis – Advent of Xpert MTB/RIF expected to increase the MDR-TB case load ≤ 70% (Meyer-Rath et al 2012)

What Can Be Done About This?

• Most important: improve drug-sensitive TB treatment to prevent

drug resistance

– First line cure rate 64-73% (WHO, South Africa TB Country Profile 2012) – 83% of patients in study sample had a history of previous TB

• New guidelines (2011) allow some MDR-TB patients to be initiated

and treated as outpatients only and shorten admission for others (until smear conversion rather than culture conversion)

• Cost reduction could be substantial, but…

– Only about half of patients (or fewer) thought eligible for outpatient initiation— must be smear-negative and in “good or fair condition” – Implementation of new guidelines has barely started – Substantial additional funding needed to implement new strategy (improved infrastructure, more outreach staff, better drugs)

A Little Back-of-the-Envelope Cost Comparison

*Costs for inpatient care and clinic visits to down-referral TB wards and primary health clinics only; drugs and lab tests and fixed costs of facilities assumed to be the same for both strategies

Inpatient model (old guidelines)

124 days inpatient care for smear +; 95 days for smear - 57 outpatient visits for smear +; 78 outpatient visits for smear - Average cost/patient $17,333*

Outpatient model (new guidelines)

42 days inpatient care for smear +; 0 days for smear - 115 outpatient visits for smear +; 132 outpatient visits for smear - Average cost/patient $4,061*

Sources: Schnippel et al 2012; Pooran et al 2013; NDOH Policy Framework 2011

But, As Usual, the Devil Is In the Details

Sources: Schnippel et al 2012; Pooran et al 2013

Value (costs exclude drugs and laboratory tests) Our back-of-the-envelope analysis Cost/inpatient initiate $7,210 Cost/outpatient initiate $911 Average cost/patient if 35% eligible $5,005 Average cost/patient if 90% eligible $1,541 Difference if 10,000 patients treated $35 million

• Very unclear how many patients can be initiated as outpatients
• Recent publication (Pooran et al 2013) assumed 90% (no evidence

provided)

• Our data suggest 35% (smear negative and BMI ≥ 18)
• What difference would it make?

The Cost and Impact

• f Scaling Up

Xpert MTB/RIF

South African M edical J

• urnal 2013; 103: 101-06

Xpert MTB/RIF

• Rapid, self-contained, cartridge-based system

for testing sputum samples for TB and for resistance to rifampicin

– High specificity and sensitivity and feasibility of routine use in laboratories – Endorsed by WHO in 2010 for TB diagnosis in high HIV prevalence countries – Diagnosis only—cannot be used to monitor treatment progress

• In March 2011, South Africa’s Minister of

Health announced national rollout of Xpert to every district by 2013

– National Health Laboratory Service responsible for rollout but had no projections of demand, procurement requirements, or cost – Asked us to estimate the incremental cost of using Xpert as first-line diagnostic for TB, compared to smear/culture

Data and Model

• Developed models of national demand for TB diagnosis and of TB

test costs under two scenarios: current guidelines (smear/culture) and Xpert

• Parameters estimated from national laboratory diagnostic

database and TB case registry

• Assumptions made about % of suspects who have TB, HIV

prevalence among suspects, rate of growth in numbers of suspects, TB treatment uptake, etc. based

• n literature and expert opinion
• Costs estimated from previous HE2RO

work on clinic costs, public sector price schedules for drugs and lab tests, and a new analysis of the cost per Xpert test

Key Results

Results by 2014 (full Xpert scale) Smear + culture Xpert scenario % change Number of suspects 2.6 million TB cases diagnosed 338,407 426,588 +26% Resistant TB cases diagnosed* 5,032 12,628 +151% % of patients diagnosed by second visit 46% 87% +89% Treatment initiated (% of diagnosed) 258,770 (66%) 359,271 (85%) +39% Cost per TB suspect tested $48 $60 +24% Cost per TB patient initiated on treatment $468 $456

• 3%

Total annual cost of first-line TB diagnosis and treatment $283 million $348 million +23%

*Includes mono and poly resistance as well as MDR-TB

Where Should Xpert Be Placed?

• Holy grail of TB diagnosis is point-of-care assay

that is accurate, rapid, easy to administer, and doesn’t cost a lot per test

• Xpert is:

– Very accurate – Somewhat rapid (2 hours) – Somewhat easy to administer (sputum sample taken and inserted in cartridge) – Cost/test depends on volume of tests performed per instrument and technician

• Decision: should Xpert be placed at centralized

laboratories or at points of care?

• Research question: How much more would it

cost to place Xpert at all points of care rather than at centralized laboratories?

Results

Variable Laboratory Clinic Number of Xpert instruments to be procured 274 4,020 Instrument procurement cost $16 million $41 million Cost per test performed $26.54 $38.91 Total annual cost of program $71 million $107 million

• Placing Xpert at point of care would

cost 51% more than at laboratories

• Laboratory placement precludes same-

day TB treatment initiation for most patients

• Will TB treatment uptake improve

enough to justify the extra cost?

• What are the implications for HIV test

and treat strategies?

$0 $10 $20 $30 $40 $50

cartridges cartridge procurement labor

• verhead

sample and supply transport instrument calibra on consumables quality ensurance instrument procurement equipment and renova ons

Point of care placement Laboratory placement

What About Xpert Negatives?

• >50% of TB suspects in South Africa have an initial negative Xpert result

– Model estimates that in 2014, 1.4 million HIV-infected TB suspects will test negative on first Xpert (53% of all suspects) – Current algorithm requires these patients to return to clinic and provide a second sputum sample for culture plus exam, x-ray, and presumptive antibiotics – 60% of TB diagnostic budget spent to find 10% of TB cases

• Estimated cost of two Xpert

tests (X/X) instead of Xpert+culture (X/C)

– If first Xpert test is negative and suspect has HIV, follow with a second Xpert test rather than culture/x-ray/antibiotics

Results

Conclusions: A TB Economics Research Agenda

Prevention

• Cost-effectiveness of education and
• utreach? (“It can’t be cost-effective if it’s not

effective.” K Freedberg)

• New(er) opportunities to prevent TB in

the community

– Changes to housing construction, public transport practices, workplace design, clinic layout, etc.—Effective but expensive? Priorities? – Active case-finding (door to door, contact tracing, etc.)—Worth doing? Sustainable? – Incentives for symptom screening or treatment completion—New approach?

• Role of HIV epidemic

– >60% of TB patients have HIV – Would more investment in ART be a cost- effective way to manage TB?

http://www.kicktb.co.za/

Diagnosis

• Monitoring and evaluation of Xpert

– Will Xpert perform well under large scale field conditions? – Improved diagnostic accuracy—how will the increase in case load affect costs? – More drug-resistant TB identified—how much will DST and treatment cost? – Linkage to care—will rapid diagnosis lead to better uptake of treatment, even if not same-day? – Benefits of rapid diagnosis—will patients and households benefit? – For other African countries, is Xpert a good investment?

• New diagnostics coming along the pipeline

– Urine LAM test for drug-sensitive TB; other options coming? – Better MDR-TB diagnostics – What combination of diagnostics and screening algorithms is feasible and affordable for different countries?

Treatment

• Drug-sensitive TB

– How much should be invested in improving cure rates? – Is direct observation of treatment cost-effective? – What are the productivity costs of TB and the benefits of treatment?

• Drug-resistant TB

– In what settings is outpatient MDR-TB treatment effective and cost-effective? – Centralized labs and facilities with transport of samples and patients, or decentralized labs and facilities with transport of instruments and providers? – Can South Africa and other African countries afford new and better drugs?

• TB/HIV

– Is “integration” of HIV and TB diagnosis and treatment cost effective?

An Unwonted(ly) Optimistic Ending

Source: Proops 2012

References

Brust J, Gandhi NR, Carrara H, Osburn G, Padayatchi N (2010) High treatment failure and default rates for patients with multidrug-resistant tuberculosis in KwaZulu-Natal, South Africa, 2000-2003. Int J Tuberc Lung Dis 14: 413–419. Directorate Drug-Resistant TB TB & HIV (2011) Multi-drug resistant tuberculosis: A policy framework on decentralised and deinstitutionalised management for South Africa. http://www.doh.gov.za/docs/policy/2011/policy_TB.pdf Farley JE, Ram M, Pan W et al (2011) Outcomes of multidrug resistant tuberculosis among a cohort of South African patients with high HIV prevalence. PLoS One 6: e20436. Fitzpatrick C, Floyd K (2012) A systematic review of the cost and cost effectiveness of treatment for multidrug-resistant tuberculosis. Pharmacoeconomics 30: 63– 80. Gandhi NR, Shah NS, Andrews JR, Vella V, Moll AP, et al. (2010) HIV coinfection in multidrug- and extensively drug-resistant tuberculosis results in high early

• mortality. Am J Resp Crit Care Med 181.

Long L, Brennan A, Fox MP, Ndibongo B, Jaffray I, et al. (2011) treatment outcomes and cost-effectiveness of shifting management of stable ART patients to nurses in South Africa: An observational cohort. PLoS Med 8: e1001055. Loveday M, Wallengren K, Voce A, Margot B, Reddy T, et al. (2012) Comparing early treatment outcomes of MDR-TB in decentralised and centralised settings in KwaZulu-Natal, South Africa. Int J TubercLung Dis 16: 209–215. Lozano R, Naghavi M, Foreman K, Lim S, Shibuya K, et al. (2012) Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: A systematic analysis for the Global Burden of Disease Study 2010. Lancet 380: 2095–2128. Meyer-Rath G, Schnippel K, Long L, Macleod W, Sanne I, et al. (2012) The impact and cost of scaling up GeneXpert MTB/RIF in South Africa. PloS One 7: e36966. Pooran A, Pieterson E, Davids M, Theron G, Dheda K (2013) What is the cost of diagnosis and management of drug resistant tuberculosis in South Africa? PloS One 8: e54587. Proops D (2012) Turning the tide: New York City’s experience in controlling tuberculosis. http://www.ggdrotterdamrijnmond.nl/fileadmin/user_upload/GGDbestanden/Hygiene/MD_PHD_Douglas_Proops_- _New_York_Citys_experience_controlling_Tuberculosis.pdf Schnippel K, Meyer-Rath G, Long L, Macleod W, Sanne I, et al. (2012) Scaling up Xpert MTB/RIF technology: the costs of laboratory- vs. clinic-based roll-out in South Africa. Trop Med Int Health 17 (9):1142-51. Schnippel K, Meyer-Rath G, Long L, Stevens W, Sanne I, et al. (2013) Diagnosing Xpert MTB/RIF negative TB: Impact and cost of alternative algorithms for South

• Africa. S Afr Med J: 103;101-106.

Schnippel K, Rosen S, Shearer K, Martinson N, Long L, et al. (2013) Costs of inpatient treatment for multi-drug-resistant tuberculosis in South Africa. Trop Med Int Health 18: 109–116. Statistics South Africa (2011) Mortality and causes of death in South Africa, 2010: Findings from death notification. http://www.statssa.gov.za/Publications/P03093/P030932010.pdf. United Nations (2011) Estimates of mid-year population: 2002-2011. United Nations Demographic Yearbook. http://unstats.un.org/unsd/demographic/products/dyb/dyb2011/Table05.pdf. World Bank (2012) World Development Indicators. http://data.worldbank.org/indicator/ World Health Organization (2010) M/XDR-TB: 2010 Global Report on Surveillance and Response. http://whqlibdoc.who.int/publications/2010/9789241599191_eng.pdf World Health Organization (2012) Global Tuberculosis Control WHO Report 2012. http://www.who.int/tb/publications/global_report/en/ World Health Organization (2012) South Africa TB Country Profile and TB Finance Profile. http://www.who.int/tb/country/data/profiles/en/

Acknowledgements

• HE2RO team in South Africa and Boston, with

special thanks to Kate Schnippel and Gesine Meyer-Rath

• USAID/South Africa
• PEPFAR
• Study sites and local collaborators
• South Africa National Department of Health and

National Health Laboratory Service

HE RO

2