Effect of directly observed treatment short

Effect of directly observed treatment short course chemotherapy for
management of tuberculosis
Dr Gulrez Shah Azhar
Master of Public Health
Copenhagen University
Year: 2010
-i-
Effect of directly observed treatment short course
chemotherapy for management of tuberculosis
Supervisors
Dr Sören Thybo
&
Dr Alan O’Rourke
Reviewer: Dr. Gulrez Shah Azhar
- ii -
Acronyms
•
AFB
Acid Fast Bacilli
•
AIDS
Acquired Immune Deficiency Syndrome
•
CASP
Critical Appraisal Skills Program
•
DOTS
Directly Observed Treatment Short course chemotherapy
•
GDP
Gross Domestic Product
•
HIV
Human Immune-deficiency Virus
•
MeSH
Medical Subject Heading
•
MDG
Millennium Development Goals
•
MDR
Multiple Drug Resistant
•
RCT
Randomised Controlled Trial
•
RNTCP
Revised National Tuberculosis Control Program
•
SCC
Short Course Chemotherapy
•
TB
Tuberculosis
•
TU
Tuberculosis Unit
•
UN
United Nations
•
WHO
World Health Organisation
•
XDR
Extremely Drug Resistant
- iii -
Abstract
Background: Tuberculosis is a leading cause of morbidity and mortality in
developing countries. DOTS is the recommended management strategy which is
followed all over the world including under the RNTCP in India. A quarter of patients
(relapse, failure & default) under RNTCP return for another round of treatment
(retreatment) which is a burden on the program. Data on relapse is not collected
under RNTCP as the patients are not followed for any period of time after completion
of treatment.
Objective: to conduct a systematic review of literature and determine evidence
regarding recurrence of TB after its successful treatment with standard short course
chemotherapy under DOTS guidelines.
Methods: Ten databases were searched including Medline, Cochrane database,
Embase and others and reference lists of articles. 255 papers resulted from these
searches. Seven studies were finally included in the review after applying the
inclusion, exclusion and quality assessment criteria.
Results & Conclusions: Relapse rate is high (almost 10%) in India which is higher
than international studies. Majority of relapse cases present soon after completion of
treatment (first six months). Risk factors for relapse included drug irregularity, initial
drug resistance, smoking and alcoholism Sex and weight were not risk factors in
India. The outcome of relapse cases put on treatment is positive but less effective
than new cases. There are sound arguments and sketchy evidence that DOTS
Category 2 treatment may not be adequate for retreatment patients.
[Word Count: 233 Words]
Keywords
DOTS, RNTCP, TB, Relapse, Reinfection, Tuberculosis, India, Systematic review
- iv -
Acknowledgements
I would like this opportunity to express my profound gratitude and heartfelt
appreciation to my supervisors Dr Sören Thybo and Dr Alan O’Rourke for their
helpful suggestions and assistance. Without their patient encouragement and
support for this thesis would not have been possible.
I wish to acknowledge all lecturers and staff of the Faculty of Health Sciences,
University of Copenhagen and School of Health And Related Research (ScHARR),
University of Sheffield for their kind support in directing my post graduate studies.
I have to mention the role of the administrative team of the Europubhealth course in
Rennes and in concerned Universities in the successful coordination of the course in
the various university campuses across different countries.
I am especially grateful to all my course colleagues for creating a friendly and fruitful
atmosphere throughout the duration of the course. I enjoyed their helpful
suggestions & criticism and will cherish their friendship even after this study program
is over.
I will always remain grateful to my family, wife and son who been very supportive to
me during this course of study and patiently waiting for me in India since the last two
years. I hope my wife and son will forgive me for the pain I have put them through
these years.
-v-
Contents
Chapter 1: Introduction & Background
1. TB
2. WHO STOP TB strategy
3. MDG’s
4. DOTS
5. TB in India
6. RNTCP
7. Relapse / recurrence
8. Aspects of healthcare system in India
Chapter 2: Methods for systematic review
1. Research question
2. Scoping search
3. Development of inclusion and exclusion criteria
a. Inclusion criteria
b. Exclusion criteria
4. Developing search terms
5. Details of each database that was searched
6. Other sources of data
a. Grey literature
b. Hand searching
c. Searching by author and institution
7. Discussion of other possible sources of data and justification for their noninclusion
8. Quality assessment
9. Data extraction
- vi -
Chapter 3: Method results
1. Characteristics and quantity of the search results
a. Preliminary search stage / scoping search
b. Main search stage
c. Number of studies included in this study at first stage
2. Results of assessment of papers against inclusion and exclusion criteria
a. Title and abstract review
b. Full text review
3. Results of assessment against quality assessment criteria
Chapter 4: Review findings
Chapter 5: Discussion
1. Use of terminology and methodology
2. Summary of study findings
Chapter 6: Conclusion and recommendations
1. Summary of main findings
2. Strengths and limitations of the study
3. Implications for future research or clinical practice
References
Appendices:
Appendix 1: Ethics declaration
Appendix 2: Search Strategy
Appendix 3: CASP Checklist
Appendix 4: Data Extraction form
- vii -
Chapter 1: Introduction & Background
1. TB
TB is among the oldest infection (more than 5000 years old) known to affect
humans. It is among the most important causes of death from a single infectious
agent and a major public health problem causing an enormous burden of disease
and economic impact especially in the developing countries.(1)
It is a chronic illness caused by Mycobacterium tuberculosis and estimated to infect
around a two billion people (a third of global population). About 10 percent of those
infected are expected to develop the disease in their lifetime. And sadly this figure
increases to 10 percent in one year for those who are HIV positive. The World Health
Organization (WHO) regards TB as a global epidemic. There were 9.27 million
incident cases of TB in 2007, the largest percentage (55%) of which was in Asia. (2)
Although TB most commonly affects the lungs but it also can involve almost any
organ of the body. The most common presenting symptoms are cough,
expectoration, fever, weight loss and haemoptysis. It can be cured by a course of
antibiotics. (3)
During the last century the prevalence of TB has been falling dramatically. This may
be due to improvement in living conditions, health care and socio economic changes.
What is most surprising is that the largest decline occurred already before antibiotic
treatment was introduced.(4)
Mismanaged treatment of TB can lead to development of Multi Drug Resistant
(MDR) TB. There are second line drugs (more side effects, expensive) to treat MDR
TB but misuse of these drugs leads to development of Extremely Drug Resistant
(XDR) TB. Because of resistance of XDR TB to both first and second line drugs
there are limited treatment options left. (5, 6)
TB is an important cause of HIV related deaths. HIV leads to a lowering of immune
status and TB develops quickly in this depleted immune state. Thus HIV positive
--1--
people are an important reservoir of TB. Since long these two were considered as
separate and the efforts were directed accordingly but now it is accepted that shared
efforts are needed to battle this dual epidemic.(7)
2. WHO STOP TB strategy
Global TB control has made great progress in the past decade. The widespread
implementation of the internationally recommended Directly Observed Treatment,
Short-course (DOTS) strategy has proved to be an effective tool in controlling TB on
a mass basis and is being practised in over 190 countries.(8)
While maintaining the current status, the prime task for the next decade is to achieve
the Millennium Development Goals (MDGs) and related Stop TB Partnership targets
for TB control. Meeting these targets requires a coherent strategy that enables
existing achievements to be sustained, effectively addresses the remaining
constraints and challenges, and underpins efforts to strengthen health systems,
alleviate poverty and advance human rights.
The new WHO Stop TB Strategy, released in 2006, has identified six principal
components to realise the global TB related MDGs by 2015. They are:(9)
•
Pursuing high quality DOTS expansion and enhancement
•
Addressing TB-HIV, MDR-TB and other challenges
•
Contributing to health system strengthening
•
Engaging all care providers
•
Empowering patients and communities
•
Enabling and promoting research
3. MDG’s
TB control is also a UN priority as mentioned in the UN Millennium development
goals.(1, 10)
Goal 6: To combat HIV/AIDS, malaria and other diseases
--2--
Target 8: To have halted by 2015 and reverse the incidence of malaria and other
major diseases, including tuberculosis
Indicator 23: Between 1990 and 2015, to halve the prevalence and death rates
associated with tuberculosis
Indicator 24: By 2005, to detect 70% of new smear positive TB cases arising
annually, and to successfully treat 85% of these cases
4. DOTS
Directly
Observed
Treatment
Short
course
(DOTS)
is
an
internationally
recommended strategy for TB control that has been recognized as a highly efficient
and cost-effective. It comprises of five elements:(9, 11, 12)
•
Sustained political and financial commitment,
•
Diagnosis by quality ensured sputum-smear microscopy.
•
Standardized short-course anti-TB treatment (SCC) given under direct and
supportive observation (DOT).
•
A regular, uninterrupted supply of high quality anti-TB drugs.
•
Standardized recording and reporting
All cases of TB under DOTS are categorised into these categories:(8)
Category 1: New sputum smear positive cases (high bacillary population)
Category 2: Retreatment
Category 3: Sputum smear negative cases (low bacillary population)
Category 4: DOTS plus for MDR TB
5. TB in India
India accounts for maximum number of cases (2.0 million). India has the highest
burden of TB of any country in the world accounting for one-fifth of the global
incidence. Every day more than 5000 people develop TB and 1000 die, that is to say
--3--
two deaths every three minutes occur in the country due to TB. Deaths due to TB
exceed the combined deaths from all other communicable diseases. TB is also the
leading killer of women, causing more orphans than those produced by all causes of
maternal mortality combined. (1, 8)
Given the largest number of cases in the country an effective TB control program in
India is essential in the global TB control efforts.
6. RNTCP
In India, Revised National Tuberculosis Control Programme (RNTCP), based on the
DOTS strategy, began as a pilot project in 1993 and was launched as a national
programme in 1997. Rapid RNTCP expansion began in late 1998 and the entire
country was covered under DOTS by 24th March 2006.
Under this program everyday more than 15,000 suspects are examined for TB free
of charge. The diagnosis of these patients and the follow-up of patients on treatment
is achieved through the examination of more than 50,000 laboratory specimens. As a
result of these examinations, each day, about 3,500 patients are started on
treatment stopping the spread of TB in the community and the country. In order to
achieve this, more than 600,000 health care workers have been trained and more
than 11,500 designated laboratory microscopy centres have been upgraded and
supplied with binocular microscopes since the inception of RNTCP.(1, 8)
The cure rate under the program is 85 percent and the case detection rate is
70percent. Most of the states in India have achieved this target.
Table 1.1: Case definitions under RNTCP(13)
Types of cases
Treatment outcomes
New: A TB patient who has never had Cured: Initially sputum smear positive
treatment for TB or has taken anti- patient who has completed treatment and
tuberculosis drugs for less than one had negative sputum smears, on two
month
occasions, one of which was at the end
--4--
of treatment.
Relapse: A TB patient who was declared
cured or treatment completed by a Treatment completed: Sputum smear
physician, but who reports back to the positive patient who has completed
health service and is now found to be treatment with negative smears at the
end of the intensive phase but none at
sputum smear positive
the end of treatment.
Transferred in: A TB patient who has
been received for treatment into a OR: Sputum smear negative TB patient
Tuberculosis Unit after starting treatment who has received a full course of
in another unit where s/he has been treatment and has not become smear
positive during or at the end of treatment.
registered
Treatment after default: A TB patient OR: Extra-pulmonary TB patient who has
who received anti-tuberculosis treatment received a full course of treatment and
for one month or more from any source has not become smear positive during or
and returns to treatment after having at the end of treatment.
defaulted, i.e., not taken anti-TB drugs
consecutively for two months or more, Died: Patient who died during the course
and is found to be sputum smear of treatment regardless of cause.
positive.
Failure: Any TB patient who is smear
Failure: Any TB patient who is smear positive at 5 months or more after
positive at 5 months or more after starting treatment. Failure also includes a
starting treatment. Failure also includes a patient who was treated with category 3
patient who was treated with Category 3 regimen
but
who
becomes
smear
regimen but who became smear positive positive during treatment.
during treatment
Defaulted: A patient who has not taken
Chronic: A TB patient who remains anti-TB drugs for 2 months or more
smear positive after completing a re- consecutively after starting treatment.
treatment regimen.
Transferred out: A person who has
--5--
Others: TB patients who do not fit into been transferred to another Tuberculosis
the above mentioned types. Reasons for Unit / District and his / her treatment
putting a patient in this type must be outcome is not known.
specified.
7. Relapse / Recurrence
In India, under RNTCP, the number of retreatment cases is 24 percent.(1) The
causes of retreatment include relapse of the disease after successful completion of
treatment, treatment failure and default in treatment. These patients already have
taken a course of anti-tubercular treatment and have been forced to return back for
another course of treatment. Information about relapse is not collected in RNTCP
and the patients are not followed up for any period of time after successful
completion of treatment. Given the high cost of treatment for each patient under
RNTCP and the potential of spread of disease from these patients, it is crucial for the
success of program and control of the disease in the country to find out more about
the reasons behind this.
Strong evidence regarding the usefulness of DOTS and more information about
relapse would need a well-designed randomised controlled trial. But this is
considered unethical by some authors given the positive experience from it in
practice settings.(14) Thus an attempt is being made in this dissertation to find out
more about these aspects of the TB control program in India by reviewing the
existing evidence from published studies by the method of systematic review.
8. Aspects of health care system in India
Health in India is a neglected subject. It is evident in the government funding of the
healthcare system. It used to be 0.9 percent of GDP (fifth lowest in the world).(15)
Majority of this amount (two-thirds) is used for the payment of salaries of the
employees. There have been numerous committees (Bhore, Kartaar Singh,
Jungalwala, Bajaj, Mudliar, Mukerjee, Srivastava, Chaddha) in more than half a
--6--
century since independence which have suggested different improvements in the
state of the health system.(16, 17)
Some of the glaring deficiencies in the public health system include massive
shortages of staff, unequal rural-urban distribution of facilities, corruption, poor
salaries and working conditions leading to endemic absenteeism, and unregulated
private sector.(18) Not surprisingly, expenditure on healthcare is the single largest
cause of falling into poverty.(19) There has been criticism that vertical programs (like
RNTCP) take away precious resources from the already dysfunctional health
system. (20)
The private health system is characterised by the variety of health care providers.
There are multiple systems recognised by the government including Ayurveda,
Unani, Siddha, Homeopathy etc.(21) There are also huge numbers of unregistered
medical providers commonly called “quacks” or “jhholachhap”.(22) These are more
commonly found in rural areas and slums. It is also very common to observe over
the counter purchase of drugs from pharmacy shop, sometimes on the advice of the
chemist (who may be the shop owner). Most of the health facilities in the private
sector are the single practitioner outpatient clinics and small inpatient nursing
homes. These usually lack adequate facilities of diagnosis and treatment.(23)
Medical education in India also suffers from bottlenecks. There is a grave shortage
of post-graduate (speciality) training programs relative to the number of
undergraduate seats.(24) A number (many thousands) of doctors prepare couple of
years to clear admission tests for these courses instead of working. This also leads
to a huge migration of doctors to developed countries for training opportunities.
There has been a gradual but significant decline in the number of students seeking
admission to medicine. This is partly also due to very low salaries offered after a
decade of studying medicine.
Thus the neglected state of health in the country, deficiencies in the system, the
sheer variety of providers and bottlenecks in the education system make it a
complex setting to understand the TB control program.
--7--
Chapter 2 Methods for the systematic review
This chapter reports the methods used in the review process. It explains the
methods used in identification of search terms and the development and refinement
of search strategies for use in various searches including the preliminary scoping
and the later final searches. Later in this chapter the development and application of
inclusion and exclusion criteria is discussed. The final part of this chapter explains
the development and application of quality assessment criteria and data extraction
tools.
Ethics approval of this study was not required as it was a secondary research, but an
ethics declaration signed by the reviewer and supervisor is attached as an appendix
(Appendix 1).
1. Research question
Is the DOTS based treatment effective in the treatment of tuberculosis?
The research question can be broken down into these objectives:
•
To conduct a systematic review of literature and determine the strength and
sufficiency of evidence regarding recurrence / relapse of TB after successful
treatment with standard six month treatment regimen
•
To discuss other factors that influence retreatment under RNTCP in India.
•
To identify key aspects of further research in this area
2. Scoping search
An initial scoping search was conducted in Ovid-MEDLINE database. Use of study
questions was made for guidance in identification of search terms. Scoping search
was intended to get an idea of the standard and quantity of available studies on this
topic. Use of scoping search was also made for identification of appropriate search
terms for the main search. The PICO concept was made use of in dividing the study
question into its four constituent components. (Table 2.1)
--8--
Table 2.1: Detailed breakdown of PICO components for this study
PICO concept
Breakdown of study question into search terms
Population
Adult age group (>18 years), developing countries, both
genders
Interventions
Successful completion of six months standardised TB
treatment
Comparisons
Non DOTS (unsupervised) treatment
Outcomes
Recurrence of TB any time after completion of treatment
The preliminary search was too broad with a huge number of results. Because the
reviewer did not have any previous experience in this type of research (systematic
review) the intention was to not miss out on any potentially relevant articles. Search
terms were enlisted for each study question (and its subparts) by using a
combination of both MeSH terms and keywords. These search terms were utilised in
searching Ovid-MEDLINE. Word truncation was also made use of to improve
sensitivity and avoid missing important papers.
As no restrictions were used in the initial scoping search this resulted in numerous.
Consequently during later searches restriction terms were added to limit the number
of results.
3. Development of inclusion and exclusion criteria
Inclusion and exclusion criteria were developed based on the research question. Use
of preliminary scoping searches was made in their testing and modification.
3.1 Inclusion criteria
The inclusion criteria which are used in this systematic review are enlisted below.
•
Study design: RCT’s, Cohort studies and other studies analysing relapse,
failure and default from India
•
Exposure: Six months standard treatment for TB
--9--
•
Outcome: Relapse / recurrence of TB any time after
•
Analysis: studies using appropriate statistical methods
•
Country of origin: India
3.2 Exclusion criteria
For a systematic review, in addition to having a clear inclusion criterion, it is vital to
have a clear exclusion criterion also. There was also a practical issue of time
constraints and size limitations for a dissertation project.
The exclusion criteria used in this systematic review are enlisted below.
•
Opposite of Inclusion criteria
•
Old cases of TB
•
No information of Bacteriological confirmation
•
Studies using inappropriate statistical methods
•
MDR-TB, TB-HIV, Category 2 and 3 treatment regimens
•
Methodologically unsound studies
Some publication types not considered suitable for this review included
•
Economic evaluations
•
Medical record reviews
•
Editorials, and
•
Journal letters
4. Developing search terms
The definition and development of search terms for the main search was done from
the research question and preliminary searches conducted earlier. Further
refinement of search terms was done after studying the articles obtained from
preliminary searches. Some search terms were also selectively eliminated. After this
process, the entire scoping search was repeated to ensure that the irrelevant results
were excluded and the important results remained. The search terms that remained
were both free text keywords and MeSH terms. Use of word truncation was done to
improve search sensitivity.
- - 10 - -
MeSH terms have been used in order to have relevant articles in search results. All
the articles indexed in Medline have descriptors. The use of such descriptors and
qualifiers makes it easy to search for the relevant articles. (25)
5. Details of each database that was searched
A large number of electronic databases were searched to ensure that relevant
articles are identified and included in this systematic review. This was attempted in
order not to miss any potentially relevant article. These included Ovid-MEDLINE,
The Cochrane library PsycINFO, CINAHL, Social sciences citation index, ASSIA,
ERIC, Embase and sociological abstracts.
An estimate was made that the relevant articles could be there in both medical and
sociological databases. Thus the chosen databases reflect an attempt to cover both
medical (clinical, nursing) and sociological sources of research.
6. Other sources of data
Other sources of data for a systematic review are commonly grey literature search,
Hand searching and searching by author and institution.
6.1 Grey literature
Grey literature is defined by the non-profit Grey literature Network Service (Greynet)
as “Information produced on all levels of government academics, business and
industry in electronic and print formats not controlled by commercial publishing i.e.
where publishing is not the primary activity of the producing body.” (Luxembourg,
1997 – Expanded in New York 2004) However there is a general agreement that
there is a need for a new conceptual framework in view of new channels of scientific
communications.(26)
Grey literature is not a particular type of document but a way of access to
information. This is considered difficult to collect due to lack of bibliographic controls
- - 11 - -
(author, publisher, date) and low print runs. But it is important in systematic reviews
(particularly which use meta-analysis) because of publication bias (studies showing
significance are more likely to be printed).
6.2 Hand searching
It is done in order to make sure that all relevant papers are included in the review. It
is especially useful in finding those reports and other publications which may have
been missed in electronic searches. The Cochrane handbook for systematic reviews
mentions it as an important part of the search process.(27)
Hand searching is a useful addition to the electronic search process because: (28,
29)
1. Uncovers relevant articles that may be poorly or inaccurately indexed or
sometimes even unindexed.
2. Allows researchers to quickly scan content from high impact journals
3. Ensure relevant studies are not over looked (28)
6.3 Searching by author and institution
This technique uses author’s name to search relevant reports. During scoping
searches, experts in a particular scientific field are identified by their reports and
publications. Further searches are made by using author’s last name and first initial.
Truncation function can also be used if author’s initials are known.
Searching by institution can be done if only a few parts are recollected like “done in
The University of Copenhagen”. Related articles can be retrieved by this method and
used along with subject, author and text word searching. (30)
7. Discussion of other sources of data and justification for non-inclusion
Ideally for a systematic review, all potential sources of data are used to ensure that
all appropriate articles have been retrieved. In practice this is difficult unless working
in a team.
- - 12 - -
Systematic reviews have grey literature search as an important part of the review
process for data sourcing. A decision not to search for grey literature was taken due
to lack of time and feasibility issues. This slight departure from the otherwise strict
review process methodology was undertaken because this review is a part of a
master’s dissertation project.
Hand searching is also an important part of the review process. It is done for
selected journals which are thought to be relevant to the aims and objectives of the
review. This was not done due to lack of time and availability of journals. But
included articles were searched for their reference lists. Application of the inclusion,
exclusion and quality assessment criteria was done for potentially relevant articles
found from these reference lists.
8. Quality assessment
It is a part of every stage of the review process. All the three stages in the study
including the study design, how it is conducted and data analysis should have
appropriate measures employed to minimise bias.(31)
For this review, checklist from CASP (Critical Appraisal Skills Program) was
employed. In the CASP checklist, there are 10 questions of which the first two are
screening questions. The remaining questions are assessed only when the response
to both of the first two questions is a yes. These screening questions are considered
as the basic questions to be answered in order to continue with the rest of the
questions in the checklist. For this study if the retrieved studies corresponded with
these two questions they were considered as potentially relevant.(32, 33) A sample
CASP checklist is attached in the appendix (Appendix 3).
A shortcoming of this systematic review is that the author is the only assessor in the
quality assessment process.
9. Data extraction
- - 13 - -
Data extraction is an important stage in the review process. The type of information
that is crucial to the study aims and objectives is pre-determined to be thoroughly
extracted from each paper. This may be a cause of bias because this is a subjective
process. An attempt was made to minimize this bias by developing a data extraction
form before the searches were done. A sample of this form was made while drawing
the study protocol. The data extraction form has enumerated the data
characteristics.(27)
The Appendix section of this review contains the data extraction form for this review
(Appendix 4). In the data extraction form, there are columns for identification of the
study, author, title, journal, publication date, etc. A summary of outcome and results
is also included in the form.
- - 14 - -
Chapter 3: Method results
This chapter details the different stages of the search process.
3.1 Quantity and characteristics of search results
3.1.1 Scoping search
As mentioned in the previous section, scoping searches were conducted on OvidMedline and other databases and potentially relevant publications were identified.
These searches were aimed to identify the correct search terms and assess the
quantity of literature available on the topic. A number of these searches were
conducted over a period of months and led to a number of studies directly and
indirectly related to the dissertation aims and objectives.
3.1.2 Main search stage
In this stage, electronic search was done and 255 papers were identified which were
considered potentially relevant. Of these, 32 publications were identified in Medline,
3 in Cochrane Database of Abstracts of Reviews of Effects, 27 in Cochrane
Database of Systematic Reviews, 12 in NHS Economic Evaluation Database in
Cochrane library, 7 in Embase, 4 in CINAHL, 30 in PsycINFO. Other databases
searched with keywords Tuberculosis and India included Social Sciences Citation
Index via Web of Knowledge (10 results), Applied Social Sciences Index and
Abstracts-ASSIA via CSA (25 results), Educational Resources Information CentreERIC via CSA (0 results) and Sociological abstracts via CSA (105 results). A
detailed search strategy employed on Medline database has been included in the
appendix (Appendix 2) as an example.
3.1.3 Total number of papers included in this study in first stage
Table 3.1 shows the selected publications from electronic searches in various
databases. The search of the 10 electronic databases led to an identification of 255
potentially relevant papers.
- - 15 - -
Table 3.1: Electronic literature identification process and search success rate
Ovid
Cochrane EMBASE PsycINFO CINAHL
MEDLINE
Total number of papers 32
3+27+12
7
30
4
2
0
0
0
4.76
0
0
0
retrieved
= 42
Number of final papers 5
selected
Success rate %
15.6
3.2 Results of assessment of papers against inclusion and exclusion criteria
3.2.1 Title and abstract review
The inclusion and exclusion criteria were applied on the results of literature search.
After title sorting for the inclusion and exclusion criteria 237 papers were rejected.
Abstract sorting of remaining 18 studies lead to a rejection of 8 more papers and so
only 10 papers were selected as relevant for further full text screening. The majority
of publications (8) came from Ovid-MEDLINE. Electronic searches in the Cochrane
library resulted in 2 more papers being selected. No relevant papers could be found
from PsycINFO, CINAHL, Embase, Social sciences citation index, ASSIA, ERIC and
sociological abstracts.
Review of papers obtained in scoping searches
In the scoping searches, there were a large number of results. These papers were
discussed with my supervisor and were helpful in planning the dissertation. However
to limit the number of results and make them manageable for a dissertation project
some limits were placed in the searches. These included studies only from India.
One paper (Hill 2002) considered relevant was excluded by these limits. This was
subsequently included in the final results. Table 3.2 lists some of the included
studies and other abstracts discussed with supervisor.
- - 16 - -
Table 3.2: Some details of included papers and abstracts reviewed and consulted
with supervisor in preliminary searches
No Study ID
Authors
1
Sisodia,
R. S. Sisodia, D. F. Wares, S. Source of
2006(34)
Sahu, L. S. Chauhan, M. Zignol
Title
retreatment cases
under the Revised National
TB
Control
Programme
in
Rajasthan, India, 2003
2
Banu
V.V. Banu Rekha, Rajeswari Assessment of long term status
Rekha
Ramachandran,
2009(35)
Rao, Fathima Rahman, A.R. Patients
K.V.
Kuppu of sputum positive pulmonary TB
Adhilakshmi, D.
successfully
treated
with short course chemotherapy
Kalaiselvi, P. Murugesan, V.
Sundaram and P.R. Narayanan
3
Vijay
Sophia
Vijay,
V.
2004(36)
Balasangameswara,
H. Treatment outcome and two &
P.
S. half years follow-up status of
Jagannatha, V. N. Saroja and New smear
P. Kumar
4
positive
patients
treated under RNTCP
Mehra
R.K.
Mehra,
V.K.
Dhingra, Study of relapse and failure
2008(37)
Aggarwal Nishi and R.P. Vashis
cases of cat i retreated with cat ii
Under RNTCP – an eleven year
follow up
5
Hill,
Hill,
A.
Ross
M.D.,
C.M.; Effectiveness
2002(14)
Manikal, Vivek M. M.D.; Riska, Observed Therapy (DOT) for
Paul F. M.D.
of
Directly
Tuberculosis: A Review
of
Multinational
Experience
Reported in 1990–2000
6
Mangura
B. Mangura, E. Napolitano, M. Directly observed therapy (DOT)
2002(38)
Passannante,
M.
Sarrel,
McDonald, K. Galanowsky,
R. is not the entire answer:
an operational cohort analysis
L. Reichman
7
Walley
John D Walley, M Amir Khan, Effectiveness
2001(39)
James N Newell, M Hussain observation component of DOTS
Khan
of
the
direct
for tuberculosis: a randomised
- - 17 - -
controlled trial in Pakistan
8
Volmink,
Volmink J, Garner P
Directly observed therapy for
2007(40)
9
treating tuberculosis (Review)
Cox,
Helen S Cox, Martha Morrow, Long term efficacy of DOTS
2008(41)
Peter W Deutschmann
regimens for tuberculosis:
systematic review
10
Davies,
G R Davies, S B Squire
Doubts about DOTS:
2008(42)
It’s too soon to say that direct
observation of short courses of
tuberculosis treatment is failing
11
Davidson,
Bruce L. Davidson
A
1998(43)
Controlled
Comparison
of
Directly Observed Therapy vs
Self-administered
Therapy
for
Active TB in Urban United States
3.2.2 Full text review
Seven potentially relevant studies were considered as included studies after the
application of inclusion and exclusion criteria. Majority of these studies originated
from Medline database and a few from the Cochrane library. Additional particulars of
these studies at this stage have been tabulated in Table 3.3.
Table 3.3: Reasons for included / excluded studies
No Study ID
1
Included reasons
Excluded reasons
Anonymous, 2001
studies
(TB
emergence
research
the
of
drug
centre,
resistance
Chennai)(44)
treatment and relapse
among
during
sputum
pulmonary
+ve
TB
patients treated with
SCC regimens
2
Chandrasekaran,
This
study
aims
2007(45)
treatment outcomes of various
- - 18 - -
to
assess
groups
of
patients.
It
also
calculates the relapse rate, time
for reporting and outcomes..
3
Cox, 2008(41)
This
study
recurrence
assesses
after
TB
successful
treatment with SCC. It included
studies from India.
4
Hill, 2002(14)
Although this study attempts to
clarify DOT’s “track record” yet it
also discusses relapse.
5
Mehra, 2008(37)
Analyses treatment outcome of
Cat 1 smear positive relapse and
failure cases.
6
Mukherjee,
Compares outcomes in sputum
2009(46)
positive TB retreatment patient
groups
(including
relapse)
at
completion of SCC under RNTCP.
7
Ormerod,
Reports
inter-
1991(47)
relationships between
relapse,
treatment
regimen
and
compliance in patients
treated
between
for
TB
1978
and
1987(before the onset
of DOTS)
8
Pardeshi,
Compares
treatment
2007(48)
outcomes in new and
retreatment
categories.
9
Santha, 1989(49)
No mention of DOTS,
Comparison between
different
- - 19 - -
regimens
(with
and
without
rifampicin) for relapse,
older study.
10
Thomas, 2005(50)
Identify risk factors associated
with
relapse
among
cured
tuberculosis (TB) patients in a
DOTS programme.
11
Volmink, 2007(40)
Although it compares DOTS with
self-administration of therapy yet it
also talks about outcomes and
relapse.
- - 20 - -
Fig 3.1: Process of identification of relevant publications
Potentially relevant articles
identified and screened for
retrieval:
N = 255
Papers rejected after reviewing
citation details:
N = 237
Total abstracts screened:
N = 18
Papers rejected at the abstract
review stage:
N=8
Total full papers screened:
N = 10
Full papers excluded:
N=3
Total full papers included in this
review:
N=7
- - 21 - -
3.3 Assessment against quality assessment criteria
All of the incorporated studies had a clearly focussed researched question. All the
included studies described their study type. The included studies were record
analysis of RNTCP program data (Chandrasekaran et al., 2007, Mehra et al., 2008
and Mukherjee et al., 2009), systematic review (Cox et al., 2008 and Volmink and
Garner, 2007), prospective observational study (Thomas et al., 2007) and literature
review (Hill et al., 2002). Details of ethics approval were not mentioned in any of
these studies. Response rates were not ascertained as most of the studies were
record based. All the studies clearly describe the results and important outcomes.
Although the all the included papers have been successfully evaluated using the
CASP checklist. But all of them have not been given equal weight. If there are any
discrepancies in the findings, more weight will be given to bigger and better designed
studies with more rigorous methodologies. The following two tables (Table 3.4 and
3.5) summarise the included studies.
Table 3.4: Main features of included studies
Study author and Study
year
Aims
& Study design / Study setting
Objectives
research
Number
of
participants
method
Chandrasekaran, Patients
2007
Cox, 2008
registered
re- Observational
All
patients 5255
were
after study, record registered for total number
default, failure or analysis
DOTS
successful
treatment
treatment
Tiruvallur
completion and
district, Tamil to my study
evaluate
Nadu,
their
treatment
may
outcome
2004
TB
All
recurrence Secondary
of patients of
in which
2608
were related
India
1999-
English 4 studies in
after successful analysis,
language
this
treatment
publications
were related
with systematic
- - 22 - -
review
standard
Hill, 2002
short review
from
3 to my study
course
electronic
chemotherapy
databases
Review
on Secondary
Manual
6822 culture
Effectiveness of analysis
search
of based
DOTS
relevant
71431 smear
studies,
based cases
reviews
and from
and
the
world.
journals.
Electronic
searches
Mehra, 2008
Treatment
Observational
All
cat
outcome of cat 1 study, record relapse
smear
positive analysis
relapse
and
cases
their
when
fate
cat
2
outcomes
TB
Observational
cat
from
2
a
chest
All
patients 234 cases of
of study, record registered for TB
positive analysis
retreatment
groups
with
clinic in Delhi
Compare
smear
TB
treated
in
regimen
Mukherjee, 2009
failure
1994 – 2005
treated
with
and patients
patients
and
failure
1 5576
TB
retreatment in
under
Bagula
RNTCP
unit,
TB
Nadia,
West Bengal
Thomas, 2005
Identify
risk Prospective
factors
study
All
patients Of 534 cured
registered for patients, 503
associated with
cat 1 DOTS were
relapse
among
treatment
TB
Tiruvallur
cured
patients
in followed up
district, Tamil
- - 23 - -
Nadu,
India
April 2000 to
December
2001
Volmink, 2007
Comparison
of Secondary
5609
Electronic
DOTS with self- analysis,
search
of participants
administered
systematic
databases,
treatment
review
manual
checking
from 11 trials
of
article
reference lists
and
contact
with
researchers
Table 3.5: Results of included studies
Study author and Main
ID
outcome Main result
Other
measure
results
Chandrasekaran, Default, failure and In period of 1999 – Of
2007
relevant
category
1
relapse in Category 1 2004 273 patients patients 64(23%) of
and category 2 under were re-registered, 273
RNTCP
of
248 76(68%)
whom
(91%)
were
112
1796
success
were re-registered
years.
and
of
re- failures, 117(6.5%)
registered within 2 of
68%
defaults,
of
failures for treatment
23%
of Median
interval
defaults from cat 1 between declaring
were
only
registered.
re- the
outcome
treatment
and
403 smear positive restart of treatment
patients were re- was 228 days for
- - 24 - -
registered,
57 default, 18 days for
were re-registered failure
twice,
8
and
212
three days for relapse
times and 1 four
times.
Cox, 2008
Recurrence
rates More
than
10% No
under DOTS from 4 recurrence
studies of India
difference
in between
observational
relapse
and recurrence in
studies carried out most studies
between
DOTS Tendency
programs in India
higher
for
recurrence
rates
among
studies
rated
as
poor
for
TB
treatment
High
degree
heterogeneity
included
of
in
studies
like
difference
between types of
patients
initially
enrolled and then
excluded
from
analysis
of
recurrence
Hill, 2002
WHO
categorisation Crude relapse rate No
of
successful was 3.6% ±2.4% in difference
treatment(completion,
cure),
in
21 culture based relapse or failure in
treatment studies
failure, death during provided
therapy, major non- treatment
adherence
significant
that prospective
post versus
follow conditions,
(default, up. 3.2% and 3.3% versus
prolonged interruption in 2 smear based settings
- - 25 - -
study
program
rural
urban
and
or loss to follow up), studies.
transfer out
developing
Quality,
versus
planned developed
Relapse rates were duration and actual countries
calculated as simple success of follow
proportion of patients
Mehra, 2008
up varied widely
of 190(3.4%) out of
Retreatment outcome 442(9%)
out
data for relapse and 4905
patients 5576
failure
cases
category
1
in
after Category 1.
of relapsed
when successful
Success
treated with category outcome.
rate
76.4% for relapse
68.5% of relapse and
2 regimen
failed
48.8%
for
reported within first failure
year. 50% of total
relapse
were
reported within first
6
months
completion
of
of
treatment
Mukherjee, 2009
following Out of 234 patients In
Outcomes
diagnosis,
in
classification
chemotherapy
according
guidelines
to
Cat
2,
unfavourable
148 outcomes
chronic
were cases were more
and (63.24%)
52 likely to be in failure
relapse,
failure group
WHO (22.22%)
(26.92%)
and 34 (14.52%) followed by default
default
(23.53) and relapse
Favourable
(6.08%)
outcome was most Age, low and hig
common in relapse grade
(76.35%)
default
and
(53.85%)
sputum
then positivity
(55.88%) sputum conversion
failure did
not
differ
significantly among
the 3 groups
- - 26 - -
and
Cure
greater
rate
was
for
grade
low
sputum
versus high grade
sputum
Thomas, 2005
Relapse
(patient Relapse rate was Drug
irregularity,
(62/503 initial
drug
cured under DOTS 12.3
%
77.4% resistance,
who has 2 sputum patients).
samples positive for (48/62)
occurred smoking
AFB
first
by
direct during
smear,1 smear & 1 months
6 alcoholism
were
of associated
with
higher likelihood of
culture positive from treatment
RNTCP relapse.
separate samples or Under
2 cultures positive)
and
case
definition
(patient who has
received
full
treatment and who
is declared cured
returns
and
is
found to have 2
positive
sputum
smear results) the
relapse rate was
10%
Volmink, 2007
Cure, completion of Retreatment
treatment,
development
participants
No
difference
who between DOTS and
of were assigned to self-administered
clinical TB (in trials of DOTS fared worse treatment in terms
drug prophylaxis)
than
those
who of cure or treatment
self-administered
treatment in one
trial
- - 27 - -
completion
Chapter 4: Review findings
Each paper selected for this review is summarised below. This summary is divided
into sections which describe the aims and objectives, study setting and time, study
methodology and sample size and the final paragraph describes the findings of the
study.
Chandrasekaran, 2007(45)
This study was designed with the objective to assess the proportion of patients reregistered after default, failure or successful treatment completion and evaluate their
treatment outcome. This study was under program conditions where the details of
patients registered under DOTS were collected from the TB register maintained in
the Tuberculosis Unit. The data was verified and keyed in twice, edited and
corrected for discrepancy and missing information.
All patients registered for DOTS treatment under the RNTCP in Tiruvallur district,
Tamil Nadu, South India in the period of May 1999 to 2004. There were 2608
patients related to my study who were started on Category 1 treatment regimen.
In the results section the study found that In period of 1999 – 2004, 273 patients
were re-registered, of whom 248 (91%) were re-registered within 2 years. Sixty eight
percent of failures and 23 percent of defaults from category 1 were only reregistered. The study also showed that many patients were re-registered multiple
times. Eg: 403 smear positive patients were re-registered, 57 were re-registered
twice, 8 three times and 1 four times. Of category 1 patients, 64 (23%) of 273
defaults, 76 (68%) of 112 failures, 117 (6.5%) of 1796 success were re-registered for
treatment. Median interval between declaring the treatment outcome and restart of
treatment was 228 days for default, 18 days for failure and 212 days for relapse.
It concluded that there was a higher default among re-registered patients resulting in
low successful treatment outcome. Only 23 percent of defaults and 68 percent of
treatment failures had re-registered for treatment. Compounding this problem is the
higher default rate of 57 percent among re-registered patients. The study stresses on
- - 28 - -
the need for motivation of re-registered patients to continue treatment. The authors
admitted that they were unable to estimate the proportion of patients re-registered
among those relapsed since they did not have the actual number of relapsed among
successful treatment completion. They stress the need for those patients who have
successfully completed treatment to report if they develop chest symptoms. The
authors also did not visit defaulted patients to find out the reasons for failure to turn
up. Also the patients could have gone to another TU or the private sector.
Cox, 2008(41)
This study aimed to find out the long term efficacy of DOTS regimen for TB by
assessing TB recurrence after successful treatment with standard short course
regimens for six months. It aimed to determine the strength and sufficiency of
evidence in support of current guidelines. This is a systematic review of evidence.
A total of 17 study arms from 16 studies were included in the review. Of these 10
were controlled clinical trials and 6 were studies either done under program
conditions or observational studies from functioning TB programs. There were four
studies from India of which three were from programs and one was a controlled
clinical trial. The three program studies (Dholakia et al, Vijay et al and Thomas et al)
had a total of 2195 (1483+178+534) patients among them, of these 2118
(1483+132+503) patients were assessed for recurrence. There were 140
(63+14+62) relapses (5.6%).
The results were that more than 10 percent recurrence rates in two of the
observational studies (Vijay et al and Thomas et al) carried out under DOTS
programs in India while the average was 5.6 percent. The authors also found no
difference between relapse and recurrence in most studies. Also, there was a
tendency for higher recurrence rates among studies rated as poor for TB treatment
High degree of heterogeneity was present in included studies like difference between
types of patients initially enrolled and then excluded from analysis of recurrence.
Exclusion criteria differed for patients with concomitant diseases and there was an
exclusion of rural patients in some studies. Definitions of relapse were sometimes
- - 29 - -
different among studies. The number of deaths which could be attributed to
recurrence was also variably assessed.
The authors discussed that despite more than a decade having elapsed and millions
being treated under DOTS the lack of published data on longer term efficacy of the
recommended treatment regimens under programmatic conditions was both
surprising and concerning. They tried to explain large variation in recurrence rates on
variation in study quality and treatment regimen used, variable inclusion and
exclusion criteria, presence of concomitant diseases, and definitions of treatment
success and recurrence. Potential contributors to recurrent TB after successful
treatment included shorter total duration of treatment (particularly rifampicin), poor
adherence during treatment (particularly intensive phase), use of less than three
drugs in intensive phase, greater disease severity and cavitation, high bacterial load,
smoking, being male, the presence of concomitant disease, being underweight and
HIV.
Hill, 2002(14)
This study aims to clarify DOTS actual track record by examining published
treatment outcomes of contemporary era cohorts. The two outcome measures were
treatment failure and post-treatment relapse. The reasoning was that constant
application of DOTS removed nonadherence from consideration and remaining
failures of a first line regimen could be attributed to biological, not behavioural,
problems.
Electronic searches led to 24 culture based and 10 smear based studies comprising
6822 and 71431 cases respectively. There were nine studies from Asia. The authors
advise caution in comparing the studies due to a broad diversity of cultural setting,
level of TB program development, clinical and mycobacteriologic methods and
prevalence of drug resistance and HIV co infection.
The results highlight good efficacy in patients who remained on therapy. The crude
relapse rate was 3.6%±2.4% in the 21 culture based studies and 3.2 and 3.3 percent
in two smear based studies. There was a substantial number of cases with short
- - 30 - -
follow up, missing data, or undiagnosed death, plus the likelihood that relapse is
more common in noncompleters (usually excluded from relapse analysis), this leads
the authors to believe that the overall relapse rate was underestimated. There is also
possibility of reinfection having falsely increased the apparent relapse rate in high
prevalence communities. Strain typing data was available for very few patients.
The authors conclude that the published experience with programmatic DOTS
(observational data) indicates that it has improved TB treatment outcomes in a wide
variety of settings. Studies in which DOTS failed to outperform unsupervised therapy
took place in settings with limited resources and suboptimal settings, indicating that
program quality must be strong for DOTS to yield its potential benefits.
Mehra, 2008(37)
This study aimed to analyse the treatment outcome of Category 1 smear positive
relapse and failure cases and their fate when treated with Category 2 regimen under
RNTCP. The reasoning behind this evidence based research is that relapses in case
of TB have same strain of organisms so are not likely to be resistant.
All patients registered under DOTS in the 17 centres covering a population of 1
million people in Gulabi Bagh chest clinic, Delhi, North India. Data pertaining to
treatment outcome of Category 1 smear positive cases from the TB register was
assessed and analysed. Relapse and failures and their time of presentation after
completion of Category 1 treatment was also recorded. They were followed up under
Category 2 treatment and their outcome recorded.
Of the 5576 patients started on treatment under Category 1, 4905 (87.9%) had a
successful treatment outcome. Of these 442 (9%) relapsed and presented on their
own to the chest clinic. According to the time of relapse, 303 (68.5%) reported within
the first year. More than half of relapses were within the first six months. Of those
patients who restarted treatment under Category 2 and could be followed up, 76.4%
had a successful outcome.
- - 31 - -
The authors discussed that due to the long follow up period of 6.4 years, the results
represent the actual state of affairs under RNTCP. Of those Category 1 relapse
patients put on Category 2, 6.2 percent fail the treatment. The authors argue that
this is due to not following the dictum “Never add one drug to a failing regimen”
(which may be very true). Such patients are only given one new drug and the
duration is increased by only 2 months. As the failure cases have still higher failure
rates, there should be a rethinking of treatment regimens to have better results.
Mukherjee, 2009(46)
The aim of this study was to compare outcomes among sputum positive TB
retreatment patient groups (relapse, failure, default) at completion of therapy under
RNTCP. The retreatment patient group (Category 2) is heterogeneous and the
subgroups are different bacteriologically and pathogenetically. But all these patients
are treated with the same regimen under the program.
It was a record based study of 234 Category 2 patients from Bagula TU, West
Bengal, West India. The TU covers a population of 0.5 million people. The time
period was between January 1999 and June 2005. .The diagnosis, classification and
chemotherapy were performed according to WHO and RNTCP guidelines. The data
was accessed from the TB register at the TU.
The results showed that 148 (63.24%) out of 234 Category 2 patients were classified
as relapse. 22.22 percent were treatment failures and 14.52 percent were defaults. A
favourable outcome was most common in relapse 76.35 percent compared to default
(55.88%) and failure (53.85%). Mean age, grade of sputum positivity and sputum
conversion did not differ significantly among the 3 groups. But cure rate was greater
for those with low grade sputum positivity compared to those with low grade sputum
positivity. Incidence of patients not cured on therapy was higher in the high grade
sputum positivity group.
The conclusion was that there was a favourable outcome in most cases of relapse.
This may be because most of the patients who relapse after a regimen to which they
were initially sensitive have bacteria that remain susceptible to the same drugs. The
- - 32 - -
authors feel that addition of parenteral streptomycin to Category 1 drugs in Category
2 should be adequate for most patients. And it should be justified to carry on this
regimen in relapse subgroup of retreatment cases. The comparatively unfavourable
outcome in chronic (failure and default) categories is due to the development of multi
drug resistance. Thus, culture and sensitivity, followed by treatment with sensitive
second line drugs be reserved for those patients in these 2 subgroups who present
with an initially high bacillary load in their sputum.
Thomas, 2005(50)
The study aims to examine the rate of relapse under program conditions
prospectively (which makes it an important study) and also the predictors of relapse
among a cohort of sputum smear positive pulmonary TB patients who successfully
completed treatment. Relapse rate, which is an important indicator of the success of
any treatment regimen, has not been measured under program conditions.
It was a prospective study conducted in Tiruvallur district, Tamil Nadu state of South
India. 17 government health centres screened symptomatic patients. The cohort was
of 715 patients between April 2000 and December 2001. Of these 534 (75%) were
declared cured and considered eligible for participation in the study. Sputum was
collected from 503 patients on a six month interval at 6 months, 12 months and 18
months and examined by fluorescent microscopy and culture. Data was collected
from multiple sources including interviews and TB register.
Results showed that the rate of relapse was 12.3 percent (62/503). The majority of
relapses, 77.4 percent (48/62) occurred during the first six months after completion
of treatment, 9 relapsed at 12 months and 5 at 18 months. On univariate analysis
drug irregularity (aOR 2.5), initial drug resistance (aOR 4.8), smoking (aOR 3.1) and
alcoholism were associated with higher likelihood of relapse. Age, sex, weight, initial
smear grade and end of intensive phase sputum conversion results did not influence
the rate of relapse.
The authors discuss the importance of regularity of treatment to ensure high cure
rates without relapse. The finding that majority of relapse occurs in the first six
- - 33 - -
months after completion of treatment is consistent with other studies. Initial drug
resistance was found to be associated with high relapse rates. But, low development
of drug resistance during therapy justifies treating patients who relapse after
treatment with currently recommended Category 2 regimen. The program
implications of this study to reduce relapse rate include ensuring that the patient
takes regular treatment and an effective counselling for quitting smoking. Proper
history of prior anti-TB treatment should be emphasised for correct categorisation of
patients for treatment.
Volmink, 2009(40)
The aim of this study was to compare DOTS with self-administration of treatment or
different DOTS options for people requiring treatment for clinically active TB or
prevention of active disease.
As this was a systematic review, electronic searches were performed, article
reference lists were checked and relevant researchers and organisations were
contacted. 11 trials with 5609 participants were selected from around the world.
Authors’ conclusions were that in low, middle and high income countries there was
no assurance that DOTS compared with self-administration of treatment has any
quantitatively important effect on cure or treatment completion in people receiving
treatment for TB. In one participating study from South Africa, there was an
increasingly negative and demoralising effect of direct observation on participants
with TB. This study found that in participants with a first episode of TB, the outcomes
were equivalent in DOTS and self-administration of treatment arms, while
retreatment participants who were assigned to DOTS fared worse than those who
self-administered treatment. But as the number of participants in the retreatment
group was small, further research is warranted to confirm these findings.
- - 34 - -
Chapter 5: Discussion
The findings of included studies which were selected (electronic literature search)
and described in previous chapters are discussed here. A narrative synthesis instead
of meta-analysis has been attempted. Reasons for not performing meta-analysis
were: firstly, due to the considerable heterogeneity of data and secondly that majority
were observational studies in this review and statistical combination of data is
usually not a prominent component in reviews of observational studies.
5.1: Use of terminology and methodology
Although under DOTS there are standard definitions (described earlier) of various
treatment categories and outcomes yet the use of these definitions and terms was
inconsistent in different studies. As an example: Thomas 2005 used a more stringent
definition of relapse than what is used in RNTCP but he considers that definition to
be less stringent than what is used in RCT’s. He defines a relapse case as a patient
cured under DOTS who has two sputum samples positive for AFB by smear, culture
or combination of both. This has a direct bearing on his results because the rate
varies, 12.3 percent according to his definition and 10 percent according to RNTCP
definition.
There were also different types of studies included in this review including
prospective, program based observational studies, literature and systematic reviews.
These studies were from different settings, the majority were from program
conditions in India and only those studies from reviews were included which had
been done in India. However considering that India is a huge country there are
considerable variations among the studies for e.g.: Mehra 2008 is from a
predominantly urban area in North India (Gulabi Bagh chest clinic, Delhi) while
Chandrasekaran 2006 is from a rural area in South India (Tiruvallur, Tamil Nadu).
This may have a bearing on results as the Category 1 relapse cases started on
Category 2 in the former are 9 percent while in the latter are 6.5 percent.
The observation period for relapse also varied in different studies. While Thomas
2005 followed up his patients for 18 months, other studies (Mehra 2008, Mukherjee
- - 35 - -
2009 and Chandrasekaran 2006) only do not specify a clear follow up period but
rather that the data analysed was from a time frame of 11, 6.5 and 5.5 years
respectively.
Anti-tubercular treatment under the RNTCP is given intermittently thrice weekly but
those seeking the treatment for the same condition outside the program are more
likely to have received a non-intermittent therapy. Although RNTCP claims that
intermittent therapy is as good as regular treatment yet there is evidence that this is
not the case.(51, 52) This is not clear in some of the studies if the treatment given is
intermittent or not. It was assumed that all studies under RNTCP were using
intermittent therapy. Some studies (Cox 2008, Hill 2002) have included all types of
therapies both intermittent and daily. Thus possible difference between intermittent
and daily regimens may be very important, however, they are very poorly
documented.
Diagnosis was smear and also culture based. While most of studies (Mehra 2008,
Mukherjee 2009 and Chandrasekaran 2006) used sputum smear microscopy for
diagnosis, some (Cox 2008 and Thomas 2005) also used sputum culture along with
smear microscopy for diagnosis. Hill 2002 separated the two into different groups
and analysed them separately.
In some of the studies it has not been made clear the differences between relapse
and reinfection. Attempts were not made to diagnose reinfection and this reinfection
may be misinterpreted as a failure to eradicate the initial strain. There is a likelihood
of this reinfection having falsely increased the apparent relapse rate especially in
high prevalence settings and HIV positive people.
5.2: Summary of study findings
In the seven studies, three were reviews (Cox 2008, Hill 2002, Volmink 2007), one
was a prospective study (Thomas 2005) and the rest three (Chandrasekaran 2006,
Mehra 2008, Mukherjee 2009) were record based studies under program conditions
of RNTCP. These seven included studies will be discussed in the following section
under two headings – relapse and other findings.
- - 36 - -
5.2.1: Relapse
As mentioned in introduction, retreatment cases constitute about 24 percent of all
cases in RNTCP. And given the high human and drug cost of treatment of each
patient (especially in the retreatment group) more information and subsequent
reduction of patients in this group is critical to the success of TB control activities.
And as the patients in RNTCP are not followed up after treatment for any length of
time there is very less information about relapse. There seems to be no trend from
year to year that the proportion of re-treatment cases becomes reduced as DOTS
has now been consolidated.
The relapse rate was high in almost all the studies. Only Thomas 2005 actively
followed up the patients and found the relapse rate to be 12.3 percent using his
stricter definition while the rate was 10 percent using the RNTCP definition. Other
researchers (Chandrasekaran 2006, Mehra 2008) reported only those patients who
had presented themselves (passively) and their relapse rates were 6.5 percent, 9
percent respectively. The obvious weakness is that the patients may have gone to
some other provider or a higher centre for treatment of relapse. Cox 2008 also found
relapse rates of more than 10 percent in two studies from India. It might be
happening that many TB re-patients (for example those coming from the private
sector and now opting for treatment in public sector) may see an advantage in not to
report that they were treated previously, so as not to be “sentenced” to eight instead
of only six months of treatment.
There were differences in relapse rates for national and international studies.
Compared to around 10 percent relapse rates in studies from India, the international
studies showed an average relapse rate of 3.6 percent in 21 culture based and 3.2
and 3.3 percent in two smear based studies (Hill 2002). Cox 2008 also found
considerable variations in recurrence rates from 0 to 14 percent in studies from
around the world.
Risk factors for relapse included drug irregularity, initial drug resistance, smoking
and alcoholism; age, sex and weight had no influence (Thomas 2005). Cox 2008
- - 37 - -
also listed potential contributors to recurrent TB as shorter total duration of treatment
(particularly rifampicin), poor adherence during treatment (mainly during intensive
phase), use of fewer than three drugs in intensive phase, greater disease severity
and cavitation, high bacterial load, smoking, being male, the presence of
concomitant disease, being underweight, and infection with HIV.
The time of presentation of relapse also varied. But the majority (68.5%) of relapse
occurred in the first year and 50 percent of total relapses were in the first six months
of completion of treatment (Mehra 2008). Thomas 2005 also found that 91.9 percent
of the 62 relapses were within the first year and 77.4 percent were within the first six
months. The period of observation in these two studies was different (11 years and
1.5 years respectively) but it is obvious that the majority of relapses occur within the
first year of successful completion of treatment. Similarly, Chandrasekaran 2006
found 91 percent of his patients were re-registered within 2 years and median
interval between declaring the treatment outcome and re-start of treatment was 212
days (approximately 7 months) for relapse patients.
The outcome of relapse patients who were put on therapy was positive in the
majority of cases. Mehra 2008 and Mukherjee 2009 report 76.4 percent and 76.35
percent of their relapse patients put on therapy as having a positive outcome (cured
or treatment completed). But this was not the case for other categories of retreatment regimen (failure and default) which had worse treatment outcomes
(discussed later). Even so the outcome of retreatment of relapse cases is poorer
than Cat.1 patient outcome,
Volmink 2007 argues that DOTS is not recommended for re-treatment patients as retreatment patients assigned to DOTS group had a worse treatment outcome than
those in self-administration of treatment group. However the sample size of the
concerned study is too small to be conclusive.
5.2.2: Other findings
On comparison of DOTS and non-DOTS based treatment of TB Volmink 2007
concludes that there is no statistically significant difference between the two
- - 38 - -
treatment groups in terms of cure or treatment completion. Hill 2002 also admits that
superiority of DOTS over unsupervised therapy for routine TB care has not yet been
shown in an evidence-based fashion. But most authorities are convinced that DOTS
improves treatment effectiveness, drug resistance rates, and overall TB control. His
contention is that it is not better only in suboptimal settings indicating that program
quality must be strong for it to yield its optimal benefits. There is also an issue of
publication bias in favour of DOTS.
Of the new smear positive patients registered under Category 1, the default and
failure rates were 12 percent and 5 percent in the study by Chandrasekaran 2006,
16 percent and 4 percent in the study by Thomas 2005. Mehra 2008 recorded a
failure rate of 3.4 percent. The distribution of default and failure cases in Category 2
patients was 22 percent and 14 percent in the study by Mukherjee 2009. Hill 2002 in
his review of studies from around the world calculated an average failure rate (unweighted mean ± SD) to be 2.4% ± 2.2% for 21 culture-based studies and 2.5% ±
1.7% for nine smear-based studies. The combined rate of failure plus default was
11.1% ± 6.7% (n = 20) and 10.0% ± 7.5% (n = 9) in culture-and smear-based
studies, respectively. These findings are comparable to findings from India.
The re-treatment outcomes for default and failure cases were not as good as
relapse. Mukherjee 2009 found 55.88 percent and 53.85 percent cases have a
favourable outcome on retreatment in the two categories respectively. Similarly
Mehra 2008 also found that only 48.8 percent cases of treatment failure on restarting
therapy had a favourable outcome. These results are similar to RNTCP treatment
cohorts.
Information about drug susceptibility was not given in most of the studies. Low drug
resistance patterns in the country suggest its minimal role in causing unfavourable
outcomes. However, the role and impact of Isoniazid resistance (which is increasing
over time (53)) may be significant, as the continuation phase of four months is
reduced to effective mono-therapy in these cases. Drug resistance was found to be
present in 20 percent cases of relapse by Thomas 2005.
- - 39 - -
Chapter 6: Conclusion and recommendations
6.1: Summary of main findings
The main findings of this systematic review are summarised below. Relapse rate is
high (almost 10%) in almost all the studies from India. It is higher than those found in
international studies. The risk factors for relapse include drug irregularity, initial drug
resistance, smoking and alcoholism. Sex and weight have been found to be
significant in international studies but not in India. Majority of relapse presented in
the first year after completion of treatment and the bulk of it occurs in the first six
months itself. The outcome of relapse patients put on treatment is positive in terms
of cure in the majority of cases but clearly less effective then the results of DOTS for
new TB cases never treated before. And there are sound arguments and some
sketchy evidence that DOTS Category 2 treatment may not be adequate for
retreatment patients.
6.2: Strengths and Limitations:
Although TB is a focus of many international activities and a number of national and
international organisations are working on its control yet the results of this systematic
review study show that there is a shortage of evidence related to the effectiveness of
DOTS versus self administered treatment as such and also a lack of precise
knowledge of the extent of relapse under program conditions in India. There are only
seven relevant publications which were found for this systematic review study. The
number of relapse cases is registered in the statistics of the RNTCP of India. Yet
there is no proper routine follow up of TB patients after end of treatment that will
allow a precise idea of the size of relapse problem.
This is one reason why there is a lack of publications which investigate this topic.
Initial scoping searches revealed a number of publications on TB but on closer
inspection most of them could not fulfil the inclusion criteria and were not included.
Compared to other studies outside India it is obvious that the choice in India of
intermittent regimen may play an important role in the creation of relapse cases
- - 40 - -
compared to daily regimens in many other countries. However, only proper well
designed prospective randomised studies may answer this query.
There is also a striking absence of good qualitative studies that are at variance to
register based studies may cast light over the preferences, adherence and felt
problems of the clients of the RNTCP. Without understanding the wishes and the
problems seen from the patients, it may be difficult to modify a strategy so strongly
advocated by the WHO and many other international organizations working with TB.
Systematic reviews of public health interventions are in themselves methodologically
challenging due to use of non-standard terminology and extensive heterogeneity.
This was the reason for performing a narrative synthesis instead of meta-analysis in
this review. There is a strong likelihood of bias in observational studies.
Undertaking a systematic review as a masters’ project is one of the limitation of this
study. Shortage of time led to this review being completed in four months only. Due
to this small timeframe study authors and publishers could not be contacted. Thus,
additional information, clarifications and missing information might not have been
available. And some papers may also have been overlooked. This would have surely
been done if this was a full systematic review and not as a part of dissertation. Hand
searching of a number of specific and relevant journals also could not be
undertaken.
Some other limitations were that I looked for evidence only in India. If evidence from
other countries was also included, international comparisons would have been very
interesting to observe. However, with some 1.2 million TB patients being treated in
India annually in the public sector under the RNTCP, India account for roughly 20
percent of global TB burden. Whatever limitation of the effectiveness of the
strategies and implementation of the RNTCP will be highly significant. Considering
that there are 15 national languages in India, use of only English language
publications is also a limitation. There is a possibility that some papers have been
missed. Also there are smaller journals which are not indexed (at all or in major
international electronic databases) and articles in these journals are likely to be
missed. Conference proceedings also could not be searched for this review.
- - 41 - -
6.3: Implications for future research and clinical practice
Lack of a common framework for data collection and analysis is an indicator for the
need of performing more studies so that the differences can be better understood
and methodologies evolved. This would make the comparisons for public health
interventions easier. The importance of relapse and its implications in successful
implementation of any TB control program using the DOTS strategy should be a
point of concern for further research. As discussed in the section above, the choice
in India of Intermittent regimen may play an important but unknown role in the
creation of relapse cases compared to daily regimens in many other countries.
However, only a proper, well designed prospective, randomized study may answer
this question. International comparisons might prove to be useful for governments for
making changes in existing strategies. Role of HIV, MDR TB re-infection with a
different strain of Mycobacterium tuberculosis and the outcomes in paediatric age
group also needs to be investigated for relapse.
[Word Count: 10 764 words]
- - 42 - -
References
1.
Central TB Division. TB India 2009 RNTCP status report. New Delhi:
Directorate General of Health Services, MOHFW GoI2009.
2.
WHO. Global Tuberculosis Control - Epidemiology, Strategy, Financing2009.
3.
Kasper D, Braunwald E, Fauci A. Harrison's Principles of Internal Medicine.
17. Auflage. New York: McGraw-Hill; 2008.
4.
McKeown T. The Role of Medicine: Dream Mirage or Nemesis1976.
5.
Drug- and multidrug-resistant tuberculosis (MDR-TB). WHO; 2010 [cited 2010
21 April]; Available from: http://www.who.int/tb/challenges/mdr/en/index.html.
6.
XDR-TB Extensively drug-resistant tuberculosis. WHO; 2010 [cited 2010 21
April]; Available from: http://www.who.int/tb/challenges/xdr/en/index.html.
7.
WHO. Tuberculosis and HIV. 2010 [cited 2010 21 April]; Available from:
http://www.who.int/hiv/topics/tb/en/index.html.
8.
Central TB Division. TB India 2010 RNTCP status report. New Delhi:
Directorate General of Health Services, MOHFW GoI2010.
9.
WHO. The STOP TB Strategy: Building on and enhancing DOTS to meet the
TB-related Millennium Development Goals2006.
10.
WHO. MDG 6: combat HIV/AIDS, malaria and other diseases. 2010 [cited
2010 21 April]; Available from:
http://www.who.int/topics/millennium_development_goals/diseases/en/index.html.
11.
WHO. Pursue high-quality DOTS expansion and enhancement. Geneva:
WHO; 2009 [cited 2009 1 November]; Five components of DOTS:]. Available from:
http://www.who.int/tb/dots/en/.
12.
Ministry of Health and Family Welfare. DOTS in India. New Delhi2009 [cited
2009 1 November]; Available from: http://www.tbcindia.org/RNTCP.asp.
13.
Central TB Division MoHFW. RNTCP at a glance. New Delhi2006.
14.
Hill A, Manikal V, Riska P. Effectiveness of directly observed therapy (DOT)
for tuberculosis: a review of multinational experience reported in 1990-2000.
Medicine2002;81(3):179.
15.
Government of India MoHaFw. National Health Policy. New Delhi2002.
16.
Bhore J, Amesur R, Banerjee A. Report of the health survey and development
committee. Volume I to1946.
- - 43 - -
17.
National Institute of Health and Family Welfare NIHFW. Committee &
Commission. 2010; Available from:
http://nihfw.org/NDC/DocumentationServices/Committe_and_commission.html.
18.
Azhar G, Jilani, AZ,,. Private providers of Healthcare in India: A policy
analysis. Internet Journal of Thirld World Medicine2009.
19.
NSSO. Report Number 441, Morbidity and Treatment of Ailments. India1998.
20.
Msuya J. Horizontal and vertical delivery of health services: what are the
tradeoffs. Background paper for the World Development Report2004.
21.
Singh P, Yadav R, Pandey A. Utilization of indigenous systems of medicine &
homoeopathy in India. Indian Journal of Medical Research2005;122(2):137-42.
22.
Dugger C. Deserted by Doctors, India’s Poor Turn to Quacks. New York
Times2004.
23.
Bhat R. Characteristics of private medical practice in India: a provider
perspective. Health Policy and Planning1999;14(1):26.
24.
National Commission on Macroeconomics and Health MoHFW, Government
of India,,. Report of the National Commission on Macroeconomics and Health. New
Delhi2005.
25.
U.S. National Library of Medicine. Introduction to MeSH - 2010. 2010 [cited
2010 21 April]; Available from: http://www.nlm.nih.gov/mesh/introduction.html.
26.
GreyNet. Grey Literature Network Service. 2010; Available from:
http://www.greynet.org/.
27.
Higgins J, Green S. Cochrane handbook for systematic reviews of
interventions Version 5.0. 0 [updated February 2008]. The Cochrane
Collaboration2008.
28.
HLWIKI. Hand Searching. 2010 [cited 2010 21 April]; Available from:
http://hlwiki.slais.ubc.ca/index.php/Hand-searching.
29.
Armstrong R, Jackson N, Doyle J, Waters E, Howes F. It's in your hands: the
value of handsearching in conducting systematic reviews of public health
interventions. Journal of Public Health2005;27(4):388.
30.
Wearn A, Rogers D. Research methods in primary care: Radcliffe Publishing;
1997.
31.
Khan K. Systematic reviews to support evidence-based medicine: how to
review and apply findings of healthcare research: Royal Society of Medicine Pr Ltd;
2003.
32.
England PHRU. Critical Appraisal Skills Programme (CASP), making sense of
evidence: 10 questions to help you make sense of reviews 2006.
- - 44 - -
33.
Oxman A, Cook D, Guyatt G. Users' guides to the medical literature: VI. How
to use an overview. Jama1994;272(17):1367.
34.
Sisodia R, Wares D, Sahu S, Chauhan L, Zignol M. Source of retreatment
cases under the revised national TB control programme in Rajasthan, India, 2003.
The International Journal of Tuberculosis and Lung Disease2006;10(12):1373-9.
35.
Rekha V, Ramachandran R, Rao K, Rahman F, Adhilakshmi A, Kalaiselvi D,
et al. Assessment of long term status of sputum positive pulmonary TB patients
successfully treated with short course chemotherapy. Indian J Tuberc2009;56:13240.
36.
Vijay S, Balasangameshwara V, Jagannatha P, Saroja V, Kumar P.
Treatment outcome and two and a half years follow-up status of new smear positive
patients treated under RNTCP. Indian Journal of Tuberculosis2004;51:199-208.
37.
Mehra RK, Dhingra VK, Nish A, Vashist RP, Nish A. Study of relapse and
failure cases of CAT I retreated with CAT II under RNTCP--an eleven year follow up.
Indian J2008 Oct;55(4):188-91.
38.
Mangura B, Napolitano E, Passannante M, Sarrel M, McDonald R,
Galanowsky K, et al. Directly observed therapy (DOT) is not the entire answer: an
operational cohort analysis. The International Journal of Tuberculosis and Lung
Disease2002;6(8):654-61.
39.
Walley J, Khan M, Newell J, Khan M. Effectiveness of the direct observation
component of DOTS for tuberculosis: a randomised controlled trial in Pakistan. The
Lancet2001;357(9257):664-9.
40.
Volmink J, Garner P. Directly observed therapy for treating tuberculosis
(Cochrane Review). The Cochrane Database of Systematic Reviews2007;2:1465858.
41.
Cox H, Morrow M, Deutschmann P. Long term efficacy of DOTS regimens for
tuberculosis: systematic review. British Medical Journal2008.
42.
Davies G, Squire S. Doubts about DOTS. British Medical
Journal2008;336(7642):457.
43.
Davidson B. A controlled comparison of directly observed therapy vs selfadministered therapy for active tuberculosis in the urban United States. CHESTCHICAGO-1998;114:1239-43.
44.
Anonymous. Low rate of emergence of drug resistance in sputum positive
patients treated with short course chemotherapy. Int J Tuberc Lung Dis2001
Jan;5(1):40-5.
45.
Chandrasekaran V, Gopi PG, Santha T, Subramani R, Narayanan PR. Status
of re-registered patients for tuberculosis treatment under DOTS programme. Indian
J. [Research Support, U.S. Gov't, Non-P.H.S.]. 2007 Jan;54(1):12-6.
- - 45 - -
46.
Mukherjee A, Sarkar A, Saha I, Biswas B, Bhattacharyya PS. Outcomes of
different subgroups of smear-positive retreatment patients under RNTCP in rural
West Bengal, India. Rural Remote Health2009 Jan-Mar;9(1):926.
47.
Ormerod LP, Prescott RJ. Inter-relations between relapses, drug regimens
and compliance with treatment in tuberculosis. Respir Med. [Research Support, NonU.S. Gov't]. 1991 May;85(3):239-42.
48.
Pardeshi GS, Deshmukh D. A comparison of treatment outcome in retreatment versus new smear positive cases of tuberculosis under RNTCP. Indian J
Public Health. [Comparative Study]. 2007 Oct-Dec;51(4):237-9.
49.
Santha T, Nazareth O, Krishnamurthy MS, Balasubramanian R, Vijayan VK,
Janardhanam B, et al. Treatment of pulmonary tuberculosis with short course
chemotherapy in south India--5-year follow up. Tubercle. [Clinical Trial
Randomized Controlled Trial]. 1989 Dec;70(4):229-34.
50.
Thomas A, Gopi PG, Santha T, Chandrasekaran V, Subramani R,
Selvakumar N, et al. Predictors of relapse among pulmonary tuberculosis patients
treated in a DOTS programme in South India. Int J Tuberc Lung Dis. [Research
Support, U.S. Gov't, Non-P.H.S.]. 2005 May;9(5):556-61.
51.
Mwandumba H, Squire S. Fully intermittent dosing with drugs for tuberculosis
in adults. The Cochrane Library2008;2.
52.
Chang K, Leung C, Yew W, Ho S, Tam C. A nested case-control study on
treatment-related risk factors for early relapse of tuberculosis. American journal of
respiratory and critical care medicine2004;170(10):1124.
53.
Tuberculosis Research Centre (ICMR) C. Trends in initial drug resistance
over three decades in a rural community in south India. Indian Journal of
TuberculosisApril 2003;50(2).
- - 46 - -
Appendix 1: Ethics Declaration
-I-
Appendix 2: Search Strategy
-1-
-2-
-3-
-4-
-5-
-6-
-7-
Appendix 3: CASP Checklist
-1-
-2-
-3-
-4-
Appendix 4: Data extraction form
No.
Study ID
Details of the paper
1
Chandrasekaran Chandrasekaran V, Gopi PG,
Santha T, Subramani R,
, 2007
Narayanan PR. Status of reregistered patients for
tuberculosis treatment under
DOTS programme. Indian J.
2007 Jan;54(1):12-6.
Observatio
2
Cox, 2010
Cox HS, Morrow M,
Deutschmann PW. Long term
efficacy of DOTS regimens
for tuberculosis: systematic
review.[see comment]. Bmj.
2008 Mar 1;336(7642):484-7.
Secondary
-1-
Methods
Objectives
Outcomes
To assess the
proportion of patients
nal study,
re-registered after
default, failure or
record
successful treatment,
analysis
completion and
evaluate their treatment
outcome
According to
standardised
RNTCP
treatment
outcome
definitions
To identify published
studies assessing
tuberculosis recurrence
after successful
treatment with
standard short course
regimens for six
months to
determine the strength
and sufficiency of
The primary
outcome was
the
proportion of
successfully
treated
patients
recorded with
recurrent
tuberculosis
analysis,
systematic
review
The
Notes
number
of
patients
There
were
2608
patients
related
to this
study
who
were
started
on
Categor
y1
treatmen
t
regimen
The 3
program
studies
from
India
had a
total of
2195
patients
among
3
Hill, 2002
evidence to
support current
guidelines.
To clarify DOT’s actual
“track record” we
review the published
treatment outcomes of
contemporary HIV era
DOT cohorts. We
examine how often and
why TB treatment was
unsuccessful
despite DOT,
during the
follow-up
period.
2 critical
outcome
measures,
treatment
failure and
post treatment
relapse
To analyse the
treatment outcome of
nal study,
Cat I smear positive
relapse and failure
record
cases and their fate
analysis
when
treated with Cat II
regimen under RNTCP
The retreatment
outcome data
for relapse
and
failure cases
of Cat I when
treated with
Cat II regimen
was reviewed.
According to
standardised
RNTCP
treatment
outcome
Hill, a. Ross; Manikal, Vivek
M.; Riska, Paul
F,Effectiveness of Directly
Observed Therapy (DOT) for
Tuberculosis: A Review of
Multinational Experience
Reported in 1990–
2000.Medicine. Issue:
Volume 81(3), May 2002, pp
179-193
Secondary
analysis
4
Mehra, 2008
Mehra RK, Dhingra VK, Nish
A, Vashist RP, Nish A. Study
of relapse and failure cases of
CAT I retreated with CAT II
under RNTCP--an eleven
year follow up. Indian J2008
Oct;55(4):188-91.
Observatio
5
Mukherjee,
Mukherjee A, Sarkar A, Saha
I, Biswas B, Bhattacharyya
PS. Outcomes of different
subgroups of smear-positive
retreatment patients under
Observatio
2009
-2-
to compare
outcomes among
nal study,
sputum-positive TB
retreatment patient
record
groups (relapse, failure
them
24
culture
based
and 10
smear
based
studies
comprisi
ng 6822
and
71431
cases
respectiv
ely
the 5576
patients
started
on
treatmen
t
234
Categor
y2
patients
6
Thomas, 2005
7
Volmink, 2009
RNTCP in rural West Bengal, analysis
India. Rural Remote
Health2009 Jan-Mar;9(1):926.
definitions
Thomas A, Gopi PG, Santha
T, Chandrasekaran V,
Subramani R, Selvakumar N,
et al. Predictors of relapse
among pulmonary
tuberculosis patients treated
in a DOTS programme in
South India. Int J Tuberc
Lung Dis. [Research Support,
U.S. Gov't, Non-P.H.S.]. 2005
May;9(5):556-61.
Volmink J, Garner P. Directly
observed therapy for treating
tuberculosis. Cochrane
Database of Systematic
Reviews 2007, Issue 4. Art.
No.: CD003343. DOI:
10.1002/14651858.CD00334
3.pub3.
and treatment after
default) at completion
of
therapy, under the
RNTCP.
Prospective To identify risk factors
associated with relapse
study
among cured
tuberculosis (TB)
patients in a DOTS
programme in South
India.
According to
standardised
RNTCP
treatment
outcome
definitions
cohort
was of
715
patients
Secondary
analysis,
systematic
review
Primary:
• Cure.
• Completion
of treatment.
• Development
of clinical TB
(in trials of
drug
prophylaxis).
11 trials
with
5609
participa
nts
To compare DOT with
self administration of
treatment or different
DOT options for people
requiring treatment for
clinically active
tuberculosis or
prevention of active
disease.
Secondary:
Keeping
outpatient
appointments.
-3-