Johns Hopkins Center for Communication Programs PatientCentered Care For the Management of Pulmonary

1
Johns Hopkins Center for Communication
Programs Patient-Centered CareFor the
Management of Pulmonary Tuberculosis March 24,
2005

• Patrick Chaulk, MD, MPH
• Annie E. Casey Foundation
• Baltimore City Chest Clinic
• Johns Hopkins Schools of Medicine and Public
  Health

2
Framework

• Global epidemiology of TB
• Clinical characteristics
• Issues related to HIV
• Management
• The challenge of treatment completion in patient
  care.
• Evidence-based approaches to assess program
  efficacy.
• Medical literature Case Studies Modeling Cost
  effectiveness analysis Quality of life
  adjustments
• Next steps and conclusions

3
Global State of Tuberculosis

• 1 new infection every minute
• 1 of the worlds population infected each year
• Globally, 1 in 3 people infected with TB
• 13 countries have 75 of all active cases
• 67 of all cases in countries that have only 7
  of the worlds GNP
• Increasing problem with drug-resistant TB
  organisms
• Single-drug resistance
• Multi-drug resistance

4
Global State of Tuberculosis

• Along with AIDS, TB is the leading cause of
  orphaned children worldwide.
• 12 million (2000)
• 20 million (2010)
• Between 2002 and 2020
• 1B people will become infected
• 36M will die

WHO Fact Sheet
5
Global State of Tuberculosis

• Problems
• Only 5 million receive some often inadequate
  treatment
• Single drug, rather than multi-drug, therapy
• Single drug added to failing regimens
• Produced major MDR outbreak in Russian prisons

Source Groups At Risk, WHO Report on TB
Elimination, 1996
6
UCSF Center for HIV Information
7
WHO DOTS Strategy

• Political commitment
• Good quality diagnosis
• Access to good quality drugs
• Systematic monitoring and accoutability
• Short course chemotherapy with supervision

8
Tuberculosis In The U.S.

• 5 per year case reduction until 1985
• Unexpected upward trend in cases
• 1985-1992 62,000 more cases than predicted
• Attributed to
• Growth in AIDS, IVDU
• Immigrants from TB endemic countries
• Poor case management and access to care
• Failure of patients to complete treatment
• Collapse of local public health/TB control
  programs
• Renewed investments in active case finding and
  efforts to enhance completion rates

9
Adult TuberculosisClinical Characteristics

• Varies widely in both immunocompetent and
  immunocompromised individuals.
• Fever, weight loss, productive cough, malaise,
  hemoptosis for weeks to months
• CXR often with apical cavitary lesion
• Sputum x 3 examined by smear and culture

10
Pediatric TuberculosisClinical Characteristics

• Clinical presentation non-specific
• Not like adult TB
• Skin testing unreliable in very young kids
• BCG
• Bacteriologic confirmation difficult
• Specimens
• Radiology variable
• Nonspecific abnormalities
• Delayed or misdiagnosis can be disastrous
• The younger the child the worse the outcome
• Vast majority of U.S. cases found through contact
  investigation

11
Treatment Completion

• Effective TB therapy exists
• 6-month, 4-drug therapy (I/R/Z/E) 95-97
  effective for susceptible organisms (HIV- and
  HIV, but is extended if cavitary lesion and is
  sputum culture at 2 months
• 6/9-month single drug therapy (I or R) 95
  effective for LTBI
• Non-completion still a challenge
• 25-30 of cases in some U.S. states still being
  treated gt1y
• 5-10 never finish full therapy
• gt80 in some developing countries

12
The Problem of Treatment Completion

• What does the scientific literature say about who
  is most at risk for non-adherence?
• All appear to be at risk Adherence is not
  predicted by a wide range of patient
  characteristics
• Adherence may be anticipated when regimens are
  long term, multi-drug, or complicatedi.e. TB
  treatment regimensat least four drugs, daily for
  6-9 months

13
The Problem of Treatment Completion

• Thus a critical question is What is the most
  effective way to improve treatment completion
  rates?
• DOT offered as a highly effective tool to
• Improve Rx completion
• Maximize cure
• Minimize drug resistance
• Reduce secondary incidence
• For everyone or just the high risk patient?

14
The Problem of Treatment Completion

• What does the scientific literature say regarding
  interventions to improve treatment completion?
• Comprehensive literature review 1966-1996
• Analysis of studies
• Interventions compared according to completion
  rates

Guidelines for Public Health Assessment of
Scientific Evidence, Feasibility and Benefits A
Report of the Guideline Development Project for
Public Health Practice, Council on Linkages
Between Academia and Public Health Practice
October 1995.
15
(No Transcript)
16
Recent RCTs South Africa

• Intervention
• Random assignment to either DO or SAT
• DO patients expected to attend clinic 5 days a
  week (x8 weeks) then THRICE a week to complete a
  6-month regimen of nurse- supervised DO
• SAT patients (or family) pick up pills once a
  week for 6 months
• Results
• DO (102) 42 cured (38)
• SAT (105) 49 cured (47)

Zwarenstein et. Al. Lancet, 19983531340-1343
17
Recent RCTs South Africa

• Comments
• High default rates in both groups
• Only 1 in 5 of the new cases included in study
• Clinic-based not community based approach
• No use of transportation and other incentives
• Workplace and school-based DOT excluded
• Without incentives SAT less burdensome for
  patients seeking care during working hours
• SAT 32 clinic visit
• DO 88 clinic visits

Zwarenstein et. Al. Lancet, 19983531340-1343
18
Recent RCTs Pakistan

• Methods
• 497 new Smear cases
• DOT by health worker 170
• DOT by family member 165
• SAT 162
• Results Treatment Completion Rates
• Health worker DOT 67
• Family DOT 55
• SAT 65

Walley et.al. Lancet 2001 357 664-669
19
Recent RCTs Pakistan

• Comments
• High default rates for all interventions
• Clinic-based program without apparent incentives
• Study launched simultaneously with DOT
  implementation
• Planning --gt (capacity building?) --gt
  Implementation
• Authors Health system an environmental
  problem?
• High cure rates are difficult within a culture
  where providers and patients expect poor service,
  which is convenient to the health system rather
  that its patients where failure is the accepted
  norm and where TB is commonly perceived to be
  incurable.
• Study site bias?
• A neighboring community had DOT cure rates 83.5
  (1998) 82.1 (1999) S cases

Walley et.al. Lancet 2001 357 664-669
20
When Is DOT Really DOT? The New Alphabet Soup

• DO
• mDOT
• DOT
• DOTs
• DOTs
• Patient centered DOT

21
Case Studies From the FieldPatient-Centered
Care

• Manila
• Faith-based model
• South Africa
• Community-based public/private model
• Harlem, NY
• Clinic-based model
• Baltimore, MD
• Community-based public health model

22
Case Studies From the FieldPatient-Centered Care

• Setting Manilla, Philipines
• Villa Sabina district of Manila (50,000) 80
  drug resistance, concentrated poverty
• Faith-based Model
• Blessed Sacrament Parish Father Sebastians
  Apostleship of Prayer
• Participation of major stakeholders
  (professional, government, academics)

Manalo, et.al. ARRD 1990142 1301-1305
23
Patient-Centered Care

• 28 volunteers Promoters
• Trained in case finding
• Did community outreach for new cases
• Follow-up with patients post hospital discharge
  for DOT
• Outcomes
• Pre-intervention completion rate 38-45
• Post-intervention completion rate 90.2
• Cost neutral
• By 2001, 65 of Philippine patients may be on DOT

Manalo, et.al. ARRD 1990142 1301-1305
24
Case Studies From the FieldPatient-Centered Care

• Setting South Africa
• Zululand 200,000 pop. miqrant labor force HIV
  36 (1993) 58 (1995)
• Community-based public private model
• Multi-site, multi-supervision approach medical
  officer, senior nurse and two field workers
  supervise health and volunteer workers
• 55 of all patients receive DOT from volunteers
• Volunteers included community residents, store
  keepers and other employers, teachers, Village
  Chiefs

Wilkinson et.al.Lancet 1994 AJPH 1996
25
Patient-Centered Care

• Treatment completion
• Pre-intervention completion rate 18
• Post-intervention completion rate
• 1991 224 88
• 1992 580 90
• 1993 674 88
• 1994 673 78
• 1994 rates dropped because of lax monthly
  auditing

Wilkinson et.al.Lancet ,1994 AJPH 1996
26
Case Studies From the FieldPatient-Centered Care

• Setting Harlem, N.Y.C.
• HIV 58, IVDU 46, 17 crack, 28 multiple
  abuse, 10 homeless, SDR 24, MDR 10, highly
  mobile
• Surrogate family model
• Individual patient assessment
• Sense of family between staff, patients
• Outreach, daily hospital visits, group
  activities, transportation, meal coupons,
  refreshments, toiletries, clothes, weekly support
  group
• Treatment completion
• Pre-intervention completion rate 11
• Post intervention completion rate 88 (99)

El-Sadr et.al. AJPH 1996861146-49.
27
Case Studies From the FieldPatient-Centered Care

• Setting Baltimore, MD
• East coast urban city with high rates of AIDS,
  IVDU, gonorrhea, syphilis, poverty, unemployment
• Community-based public health model (David
  Glasser--1975-1988)
• Phase 1 Aggressive case finding (1975-1978)
• Phase 2 Incremental implementation of DOT
• 1978-1981-Clinic-based DOT
• On average 25 of all Baltimore cases
• 1982-1992- Community-based DOT
• Range 55-70 of all Baltimore cases
• 1992-Present
• On average 90 of all Baltimore cases

28
Patient-Centered Care

• Program performance measures
• Clinical Outcomes
• Treatment completion
• Pre-intervention completion rate 55
• Post-intervention completion rate90 (DOT)
  versus 53year (SAT),1985-1992 (DOT 95-96,
  1995-1997)
• Clinical error rates 38 pvt sector v. 5.2 DOT
• DOT Relapse rates 3-6/year
• DOT managed cases with 4-fold lower mortality
  rate during therapy than non-DOT managed cases
• Lower mortality for DOT-managed patients with HIV
  infection

29
Patient-Centered Care

• Bacteriological Outcomes
• Sputum conversion to negative at 3 months
• DOT cases 2-fold higher than SAT cases
• Drug resistance extremely low (5 MDRTB (at least
  I/R)cases between 1989 and 1999)
• Epidemiological Outcomes
• TB incidence fell 67 (1978-1996)
• City rank for TB fell from 2nd to 23rd (1978-1992)

30
Patient-Centered Care

• Economic outcomes
• CEA 15,003 (SAT) v. 13,925 (DOT)/1000 Pts.
• Deaths 13 v 3 Relapse 133 v 31
• Treatment savings of 24 million, 1978-92
• Political outcomes
• Sustained through the terms of
• 4 governors
• 4 mayors
• 4 health commissioners
• substantial cuts in federal TB control

31
1
Community-based DOT
5
10
Clinic-based DOT
15
20
25
30
32
Cost Effectiveness of Patient-Centered DOT

• Baltimore Community outreach DOT program
• DOT for 90of all Baltimore patients
• Hospital case recruitment
• Community outreach home, school, work
• Universal DOT and Fixed-dose combination (FDC)
  pills more cost effective than SAT
• Decision tree model for 1000 patients (90 v
  60)
• Costs labor, travel, hospitalization, clinic and
  provider services probabilities from literature
• 13,925 (DOT), 13,959 (FDC) 15,003 3 v 9 v 13
  deaths 31 v 96 v 132 relapses
• Threshold analysis robust

Moore, et al. Am J Resp Crit Care Med
19961541013-9
33
How Cost Effective Are The Other Patient-Centered
DOT Programs?

• We dont know because they have not been formally
  analyzed, but
• Effectiveness
• Many settings with very low completion rates on
  SAT would likely be even more cost effective
  using either DOT or FDC pills
• Costs
• Strengthened by higher probability that MDRTB
  would emerge when unsupervised in these same
  settings
• Overall
• DOT or FDC should prove highly cost effective

34
Pitfalls In Cost-Effectiveness Analysis

• CEA far from silver bullet
• CEA only one perspective
• Limitations regarding public health
• TB control entails multiple interventions
• Effective TB control involves multiple
  interventions case finding, treatment,
  screening, prevention, surveillance, laboratory
  confirmation
• CEA may oversell one of these components
• CEA may not be feasible
• CEA miss other significant benefits of DOT

35
Quality of Life Benefits ofPatient-Centered DOT
36
Quality of Life Benefits Of Patient-Centered DOT
37
Traditional Self-supervised Model
TB Clinic
TB Clinic
TB Clinic
Client
Client
Client
At 2 Months
At 4 Months
At 6 Months
TB Clinic
Primary care
Primary care
Primary care
TB Clinic
TB Clinic
WIC
Social
Client
Client
Client
Social
Sputum collection
Work
Sputum Collection
22 home visits
16 more home visits
16 more home visits
Patient-Centered DOT
38
Features of Highly Effective TB Treatment
Programs

• Asset-based
• Community resources
• Patient participation in design phase
• Personalized care
• Credible intermediaries
• Messenger more important than message?
• Respectful and culturally competent
• Broadly integrated
• Diversity of stakeholders
• Decentralized with good management / auditing

39
Next Steps

• Apply what we know works
• Replicate models and best practices
• Provide customized technical assistance
• Peer-to-peer approaches
• Focus on things that are modifiable
• Field test new and refined models
• TB clubs, cross-train staff
• Create broad partnerships and collaboratives
• Tap into system reforms as TB opportunities
• Refine asset mapping and improve
  cost-effectiveness
• All communities have some assets
• Residents, existing capacity, CBOs/NGOs

40
Next Steps

• Support greater bi-national/multi-national
  partnerships
• Failure of industrialized nations to aggressively
  invest in a coordinated global TB effort will
  prove politically foolish and epidemiologically
  naïve.

41
Next Steps

• Significant role for communication
• Create clear, effective and targeted
  communication strategies about what constitutes
  culturally effective treatment strategies using
  DOT.
• Create culturally effective messages around TB
  care
• Major issue of stigma so communication can play
  an important role in health seeking behaviors,
  public health and private provider messages,

42
Conclusions

• Will DOT eliminate TB?
• No
• Will we eliminate TB without DOT?
• No
• Will DOT always lead to reductions in TB?
• No
• Is DOT cost effective?
• Probably

43
Conclusions

• So what is the role of DOT
• DOT appears to be the most consistently effective
  tool for producing the best clinical outcomes and
  highest quality of care for the most TB patients
  regardless of socio-economic setting.

44
We often measure things that mean very little,
but fail to measure things that mean the most.
Albert Einstein