Balewgizie Sileshi

ADDIS ABABA UNIVERSITY
COLLEGE OF HEALTH SCIENCES
SCHOOL OF PUBLIC HEALTH
PREDICTORS OF MORTALITY AMONG TB-HIV CO-INFECTED
PATIENTS DURING TB TREATMENT IN BAHIRDAR TOWN
GOVERNMENTAL HEALTH INSTITUTION, ETHIOPIA.
A RETROSPECTIVE COHORT STUDY
BY
BALEWGIZIE SILESHI (BSc. in PH)
ADVISOR
DR.NEGUSSIE DEYESSA (MD, MPH, PHD)
A THESIS SUBMITTED TO THE SCHOOL OF GRADUTE

STUDIES OF ADDIS ABABA UNIVERSITY IN PARTIAL
FULFILLMENT OF THE REQUIRMENTS FOR THE DEGREE OF
MASTER OF PUBLIC HEALTH IN EPIDEMIOLOGY
MAY, 2012
ADDIS ABABA, ETHIOPIA

ADDIS ABABA UNIVERSITY
SCHOOL OF GRADUATE STUDIES
PREDICTORS OF MORTALITY AMONG TB-HIV CO-INFECTED
PATIENTS DURING TB TREATMENT IN BAHIRDAR TOWN
GOVERNMENTAL HEALTH INSTITUTION, ETHIOPIA.
A RETROSPECTIVE COHORT STUDY
BY
BALEWGIZIE SILESHI
School of Public Health, College of Health Sciences
Addis Ababa University
Approved by the Examining Board
_________________________________________ ___________
Chairman, SPH signature
Dr. Nigussie Deyessa
Advisor Signature
__________________________________________ ___________
External Examiner Signature
__________________________________________ ___________
Internal Examiner Signature

ACKNOWLEDGMENTS
First of all my deepest and sincere gratitude goes to my advisor Dr. Nigussie Deyessa
for his unreserved encouragement, provision of relevant and timely comments and
guidance throughout my thesis work.
My appreciation also extended to Dr. Muluken Melese and Dr. Belayneh Girma from
MSH/HEAL TB Ethiopia and Dr.Alemayehu Worku from Addis Ababa University
School of Public Health for their unreserved and valuable comments during
development of this report.
I gratefully acknowledge and appreciate Officer of TB-HIV collaboration activities in

Amhara Regional Health Bureau, ART clinic staffs, health team leaders and Data clerk
and adherence supporter staffs for their invaluable co-operation during data collection.
Lastly I thank the U.S. Agency for International Development, MSH/HEAL TB

Project, Ethiopia for funding this study.
i
TABLE OF CONTENTS
Contents Page
ACKNOWLEDGMENTS ............................................................................................. i
TABLE OF CONTENTS ............................................................................................. ii
LIST OF TABLES ...................................................................................................... iv
LIST OF FIGURES ......................................................................................................v
ACRONYMS AND ABBREVATIONS ..................................................................... vi
ABSTRACT .............................................................................................................. vii
1. BACKGROUND ......................................................................................................1
2. LITRATURE REVIEW ...........................................................................................3
2.1. General burden of TB-HIV co-infection .................................................................3
2.2. Global and national consideration in response to the dual epidemic ........................4
2.3. Magnitude of mortality and Survival time due to TB-HIV co-infection ..................5
2.4. Predictors of mortality among TB-HIV co-infected patients ...................................5
3. OBJECTIVES ...........................................................................................................8
4. METHODS AND MATERIALS...............................................................................9
4.1. Study area and period .............................................................................................9
4.2. Study design...........................................................................................................9
4.3. Population ..............................................................................................................9
4.4. Sample size determination .................................................................................... 10
4.5. Sampling Procedures ............................................................................................ 11
4.6. Data collection procedures ................................................................................... 11
4.7. Measurement of variables..................................................................................... 13
4.8. Data analysis procedures ...................................................................................... 13
4.9. Ethical Considerations .......................................................................................... 14
5. RESULTS ............................................................................................................... 15
5.1. Enrollment of TB-HIV co-infected patients in the study ....................................... 15
5.2. Baseline Socio-Demographic Characteristics of the Study Subjects ...................... 16
5.3. Clinical and immunological characteristics of the Study Subjects ......................... 18
5.4. Characteristics of study subjects on type of TB diagnosis, type and medication .... 20
5.5. Base line social condition ..................................................................................... 20
ii
5.6. Comparison of mortality between On ART and Non-ART cohorts ....................... 22
5.7. Predictors of mortality in TB-HIV co-infected patients during TB treatment ........ 27
6. DISCUSSION ......................................................................................................... 33
7. STRENGTH AND LIMITATIONS ........................................................................ 37
8. CONCLUSIONS.................................................................................................... 38
9. RECOMMENDATIONS ........................................................................................ 39
10. REFERENCES ..................................................................................................... 40
11. ANEXES .............................................................................................................. 45
ANEX I Data collection format ................................................................................... 45
Annex II. Conceptual framework ................................................................................ 54
iii
LIST OF TABLES
Table 1: Baseline socio-demographic characteristics of TB-HIV co- infected patients

during TB treatment in Bahirdar town governmental health institutions, 2012 ............ 17
Table 2: Baseline clinical and immunological characteristics of TB-HIV co- infected

patients during TB treatment in Bahirdar town governmental health institutions, 201219
Table 3: Type of TB diagnosis and medication of TB-HIV co-infected patients during

TB treatment in Bahirdar town governmental health institutions, 2012 ....................... 21
Table 4: Baseline social condition of TB-HIV co- infected patients during TB treatment
in Bahirdar town governmental health institutions, 2012 ............................................ 22
Table 5: Actuarial life table analysis of TB-HIV co- infected patients during TB
treatment in Bahirdar town governmental health institutions, 2012 ............................. 23
Table 6: Medication and TB location associated predictors of mortality among TB-HIV

co-infected patients during TB treatment in Bahirdar town governmental health
institutions, 2012........................................................................................................ 28
Table 7: Clinical and immunological associated predictors of mortality among TB-HIV

institutions, 2012........................................................................................................ 29
Table 8: Socio-demography associated predictors of mortality among TB-HIV co-

infected patients during TB treatment in Bahirdar town governmental health
institutions, 2012........................................................................................................ 30
Table 9: Multivariate analysis of variables associated with mortality among TB-HIV

institutions, 2012........................................................................................................ 32
iv
LIST OF FIGURES
Figure 1: Enrollment of TB-HIV co-infected patients in the study, 2012 .................... 15
Figure 2: Kaplan-Meier estimate of survival among TB-HIV co-infected patients during

TB treatment with and without ART in Bahirdar town governmental health institution,
2012........................................................................................................................... 24
Figure 3: Kaplan-Meier estimate of mortality among TB-HIV co-infected patients

during TB treatment with and without ART in Bahirdar town governmental health
institution, 2012 ......................................................................................................... 24
Figure 4: Kaplan-Meier curve of mortality among TB-HIV co-infected patients during

TB treatment with and without ART among smear positive PTB patients in Bahirdar
town governmental health institutions, 2012 .............................................................. 25

TB treatment with and without ART among smear negative PTB patients in Bahirdar
town governmental health institutions, 2012 .............................................................. 25

during TB treatment with and without ART among extra PTB patients in Bahirdar town
governmental health institutions, 2012 ....................................................................... 26

during TB treatment with and without ART among CD4 level 0-75cells/µL patients in
Bahirdar town governmental health institutions, 2012 ................................................ 26
v
ACRONYMS AND ABBREVATIONS
AIDS Acquired Immune Deficiency Syndrome

AOR Adjusted Odd Ratio
ART Antiretroviral Therapy
AZT Zidovudine
CI Confidence Interval
CPT Cotrimoxazole Prophylactic Therapy
d4t Stavudine
df Degree of Freedom
EFV Efavirenz
HAART Highly Active Antiretroviral Therapy
HIV Human Immunodeficiency Virus
HR Hazard Ratio
IQR Inter Quartile Range
NNRTI Non-Nucleoside Reverse Transcriptase Inhibitor
NRTI Nucleoside Reverse Transcriptase Inhibitor
NVP Nevirapine
OR Odd Ratio
PCP Pneumocystis Carini Pneumonia
PI Protease Inhibitor
PMO Person Month Observation
PTB Pulmonary Tuberculosis
SNNPR Southern Nations Nationalities Peoples Region
TB Tuberculosis
3TC Lamivudine
TLC Total Lymphocyte Count
WHO World Health Organization
vi
ABSTRACT
Background: The synergy between tuberculosis and human immunodeficiency virus is

strong in high human immunodeficiency virus prevalence population, in which
tuberculosis is a leading cause of mortality and human immunodeficiency virus is
driving the tuberculosis epidemic in many countries especially in sub-Saharan Africa.
Objective: To assess predictors of mortality among TB-HIV co-infected patients

during tuberculosis treatment in Bahirdar town governmental health institutions.
Methods and Materials: Institution based retrospective cohort study was conducted
between April, 2009 and January, 2012. Based on TB, Pre-ART and ART registration
log book records of TB-HIV co-infected patients were categorized into On ART and
Non-ART cohorts. Chi-square test and T- test were used to compare categorical and
continuous variables between the two groups, respectively. Kaplan-Meier test was used
to estimate the probability of death after tuberculosis diagnosis. The log-rank test was
used to compare overall mortality between the two groups. The Cox proportional
hazard model was used to determine factors associated with death after tuberculosis
diagnosis.
Results: A total of 422 TB-HIV co-infected patients (272 On ART and 150 Non-ART
cohorts) were followed retrospectively for a median of 197 days (IQR: 140-221) in On
ART and 191 days (IQR: 65.5-209.5) in Non-ART cohort. In Non-ART cohort more
TB-HIV co-infected patients died during TB treatment; 44(29.3%) than On ART cohort
49(18.0%) with incidence rate of 6.03Per 100PMO (95%CI: 4.5, 8.1) and 3.20 per
100PMO (95% CI: 2.40, 4.20). Receiving ART (AHR=0.35 (0.19-0.64)), not initiation
of CPT (AHR=3.03 (1.58-5.79)), being ambulatory (AHR=2.10 (1.22-3.62)), CD4
count category 0-75cells/µl, 75-150cells/µl and 150-250cells/µl (AHR=4.83 (1.98-
11.77), 3.57 (1.48-8.61) and 3.07 (1.33-7.07) respectively) and treatment in hospital
(AHR=2.64 (1.51-4.62)) were independent predictors of mortality during TB treatment.
Conclusions: Despite the availability of free ART from health institutions, mortality
was high and was strongly associated with the absence of ART during TB treatment. In
line with other studies risk of death was over 65% lower in TB-HIV co-infected
patients treated with ART compared with those not treated with ART.
vii
1. BACKGROUND
The human immunodeficiency virus (HIV) pandemic presents a massive challenge to

the control of tuberculosis (TB) at all levels. This unprecedented scale of the epidemic
of HIV-related tuberculosis demands concerted and urgent action. The synergy between
TB and HIV is strong i.e., in high HIV prevalence population, TB is a leading cause of
morbidity and mortality, and HIV is driving the tuberculosis epidemic in many
countries, especially in sub-Saharan Africa. The spread of TB and HIV is aggravated
by socio-economic factors such as poverty and low levels of literacy(1). TB is often the
first clinical indication that a person has underlying HIV infection and TB services can
be an extremely important entry point to HIV prevention, care and treatment (2).
The syndemic interaction between the human immune deficiency virus (HIV) and
tuberculosis (TB) epidemics has had deadly consequences around the world. The TB-
HIV syndemic has had a major impact on human health and disproportionately affects
people in Africa (3).
In patients with advanced acquired immune deficiency syndrome (AIDS) and active
TB, highly active antiretroviral therapy (HAART) may be administered concurrently
with the TB treatment because if antiretroviral therapy (ART) is deferred another
opportunistic infection may superimpose and accelerate HIV disease progression
(4).The World Health Organization (WHO) currently recommends that all HIV infected
TB patients receive co-trimoxazole and that patients with clinical evidence of AIDS or
CD4+ T-lymphocyte counts (CD4) <350 cells/μL initiate ART during TB treatment
(5). ART should be initiated for all people living with HIV with active TB disease
irrespective of CD4 cell count (6).
Despite international recommendations and the proven benefit of ART, physicians

remain reluctant to prescribe ART to HIV-infected TB patients, because of concerns
about overlapping toxicity, drug-drug interactions, pill burden, and immune
reconstitution inflammatory syndrome (IRIS) (7).
Major research questions also remain unanswered, including the optimum time to
initiate ART, the optimum ART regimen in HIV-infected TB patients, and the added
1
value to ART of co-trimoxazole and other opportunistic infection prophylaxis
medications in reducing mortality during TB treatment (8).Though, mortality rate from
HIV associated TB in developing countries is high, it is not clear whether it is due to
failure of anti TB treatment or complications of HIV (9).
To date, there have been inadequate data on predictors of mortality among TB-HIV co-
infected patients in Ethiopia. The aim of this study was to answer some of the above
questions directly or indirectly and fill the information gap by assessing the predictors
of mortality among TB-HIV co-infected patients. And also it provided empirical
evidence for TB-HIV program planners, decision makers and implementers at the
different level by enabling them to access a base line data on risk factors of mortality
on TB-HIV co-infected patients. The study also aimed to compare the survival rate
among TB-HIV co-infected patients who received and did not receive ART.
2
2. LITRATURE REVIEW
2.1. General burden of TB-HIV co-infection
Worldwide 33.3 million people live with HIV and 1.8 million deaths occur due to
AIDS related illnesses. Sub-Saharan Africa still bears an inordinate share of the global
HIV burden accounted for 68% of HIV infections worldwide and 72% of the world’s
AIDS-related deaths in 2009 (10).
The annual risk of developing TB in people living with HIV/AIDS (PLWHA) who are
co-infected with M. tuberculosis ranges from 5 to 15% as compared to 5 to 10% life
time risk for HIV negative individuals. It also increases the likelihood of re-infections
and relapses of TB (1). In 2010, there were an estimated 8.8 million incident and 12
million prevalent cases of TB globally. Approximately 1.4 million people died of TB in
2009. Of the 8.8 million incident TB cases in 2010, an estimated 1.1 million (13%)
were HIV-positive with 0.35 million deaths. Of these HIV-positive TB cases,
approximately 80% were in the African Region (11).TB is a leading cause of morbidity
and mortality in patients with HIV/AIDS (1, 5, 12). Sub-Saharan Africa carries the
overwhelming share of the global burden of HIV/AIDS and of HIV-associated TB (13).
In Ethiopia 1.2million people live with HIV causing for about 28,073 annual AIDS
deaths (14). Ethiopia stands 7th place in the global rank by estimated number of cases of
TB with incidence of 261 per 100,000 populations for all forms of TB and case
detection rate of 72. The prevalence of Tuberculosis of all forms in the same period is
estimated at 394 per 100,000 populations (11). About 40-70% of HIV patients in
Ethiopia are co-infected with TB (15).
A total of 56,040 TB patients were tested for HIV in 2009, of which 11,118 (20%) were
found to be HIV positive. In addition, a total of 24,112 HIV-positive people were
referred from chronic HIV and ART clinics for TB screening out of which 4,154
(17.2%) were found to have active TB (16).
TB-HIV co-infected patients had a lower quality of life in all domains as compared to
HIV infected patients without active TB (17).
3
2.2. Global and national consideration in response to the dual epidemic
Tuberculosis and HIV programs should collaborate closely with general healthcare
providers to improve the quality and continuity of care of HIV-infected TB patients. As
well as ensuring effective TB treatment and promoting access to ART, improving the
quality of care of HIV infected TB patients involves preventing and treating common
HIV-related infections, avoiding unwanted drug effects and avoiding over-crowding of
hospital wards. The impact of HIV on TB patients and programs has implications for
TB control policies (13).
ART is recommended in all HIV-infected TB patients with CD4 counts <200

cells/mm3 and should be considered for patients with CD4 counts <350 cells/mm3. In
cases where CD4 cell counts cannot be assessed, a total lymphocyte count of
1200/mm3 or below can be used as a substitute indication for treatment in the presence
of symptomatic HIV disease. In the absence of CD4 counts and total lymphocyte
counts, which may be the case in most of sub Saharan Africa, ART is recommended for
all TB patients with documented HIV infection (18).
Current international guidelines also recommend that patients with advanced immune-
suppression begin ART within two weeks to two months of beginning TB treatment
(19).
After the establishment of TB-HIV Advisory Committee in 2002 key collaborative TB-
HIV activities being implemented by Ethiopia includes routine HIV testing and
counseling for TB patients and suspects, Co-trimoxazole Preventive Therapy (CPT),
HIV prevention services at TB clinic, HIV care and support, Provide ART for eligible
TB-HIV co-infected patients, routine offer of TB screening for all HIV positive clients,
INH preventive therapy (IPT) and TB infection control in health care and congregated
setting (1).
In Ethiopia from July 2008 through June 2009, 68% of HIV-positive TB patients were
put on Cotrimexazole Prophylaxis Therapy (CPT) and 41% of HIV-positive TB
patients have started ART (16).
4
2.3. Magnitude of mortality and Survival time due to TB-HIV co-infection
Studies showed that the rate of death among TB-HIV co-infected patients prior to
successful completion of TB treatment ranges from 8.5% to 29% (4, 20-22). In a meta-
analysis study the pooled percentage of deaths during TB treatment was 3.5% (95% CI:
2.5%–7.2%) in HIV uninfected persons and 18.8% (95% CI: 14.8%–22.8%) in HIV
infected persons (22). Death occurred in 43 of 380 (11.3%) patients exposed to ART
during TB treatment, compared with 87 of 354 patients (24.6%) without ART (RR
0.46, 95% CI 0.33–0.64) (23).
A retrospective study in Thailand found that the overall mortality rate was 7.7% in
ART+ group and 67.7% in ART- group (P < 0.001) (24). A prospective, multi-center
observational study of HIV-infected TB patients in Thailand showed that patients who
took ART had one-fifth the risk of dying than those who did not take ART (25). Many
of the adverse events observed occurred within the first 2 months of starting TB therapy
[hepatitis (91%), rash (85%), GI disturbance (79%), and peripheral neuropathy (65%),
other neurological (65%)] (4). The median time from treatment initiation to death was
51 days (range 0–443 days) (26).
Another study in northern Thailand showed that HIV-positive patients are more likely
to die in the first month of TB treatment (AOR 3.24, CI 2.46–4.26) than HIV-negative
patients. Their observation of surveillance data over 12 years revealed that the highest
proportion of TB deaths happened during the first month of TB treatment (27).
2.4. Predictors of mortality among TB-HIV co-infected patients
2.4.1. Socio-demographic risk factors
Lower median survival times were observed in the 40-49 year age bracket and in males
(45.4 and 51.5 months, respectively). Survival time was visibly reduced among patients
with no schooling (41.6 months) (28). Factors which remained independently
associated with death in the multivariate model include increasing age HR 1.02 (95%
CI: 1.00–1.04) (26). In multivariable survival analysis, age greater than 45 years (HR
1.44, 95% CI 0.96–2.15) was independently associated with long-term survival after
TB diagnosis (21). HIV-infected female patient had a lower risk of death than males
5
(adjusted relative risk: 0.55, 95% CI 0.37–0.83) (23). A study in Ethiopia showed that
age over 45 years at presentation (HR [95%CI] =1.7 [1.2, 2.4]) and men had higher
mortality rates than women (HR [95%CI] =1.4 [1.1, 1.8]) (29).
2.4.2. Base line clinical and lab risk factors
Factors which remained independently associated with death include CD4 10-24 HR
3.4 (95% CI: 2.0–6.1), CD4 25-49 HR 3.0 (95% CI: 1.6–5.7), CD4 50-99 HR 2.6 (95%
CI: 1.4–4.8) or missing CD4 count HR 3.8 (95% CI: 1.7–8.2) (26). In another study
baseline CD4 count was independently associated with long-term survival after TB
diagnosis (HR 0.76 per 100 cell increase, 95% CI 0.65–0.88) (21). In contrary study
done in south India showed that lower CD4 count of 200/mm3 indicating advanced
immunosuppressant was not associated with mortality (30). TLC<= 750/mcL at
baseline was the strongest predictor of overall mortality [HR (95%CI) = 3.4 (1.5–7.6)
(31). Higher mortality was associated with advanced clinical stage [95%CI] =2.4 [1.5,
4]) and WHO stage IV [HR (95%CI) = 9.2 (1.8–45.8), P = 0.007] at start of treatment
(29, 31).
2.4.3. TB location site, type and category risk factors
Study done in south India showed that disease classification as ‘pulmonary TB’ [(OR
2.21, CI (1.26–3.88)], type as ‘retreatment cases’ [OR 2.05, CI (1.01–4.18)] and
‘treatment irregularity’ [OR 2.07, CI (1.14–3.78)] were potential risk factors for death
(30). The other risk factors associated with a higher probability of death were
gastrointestinal TB (HR 9.2, 95% CI: 1.10-78.02, P = 0.042), multidrug-resistant TB
(HR 2.0, 95% CI: 1.04-3.78, P = 0.038) and ineffective TB regimen (HR 5.0, 95% CI
2.0–12.6) (20, 24). TB-HIV co-infected patients with mixed clinical presentation, extra
pulmonary and treatment dropout have reduced survival time (22.2, 39.6 and 45.4
months, respectively) (28).
Diagnosis of mixed TB (HR 1.4, 95% CI: 1.0–2.0) and multidrug-resistant TB (HR 3.3,
95% CI: 1.2–9.2) were factors which remained independently associated with death
(26). Having pulmonary TB was also independently associated with a greater risk of
6
death, and there was a trend towards greater risk of death in those with prior anti-TB
treatment (23).
2.4.4. Medication and regimen type risk factors
In the logistic regression analysis the treatment related factors independently associated
with death were ‘non initiation of ART’ [AOR-2.80, CI (1.15–6.81)] and of ‘CPT’
[AOR-3.46, CI (1.47–8.14)] (30). A prospective, multi-center observational study of
HIV-infected TB patients in Thailand showed that the risk of dying was further reduced
with early ART initiation and concomitant use of fluconazole. The choice of non-
nucleoside reverse transcriptase inhibitor (NNRTI) in the ART regimen did not alter the
risk of adverse events or death (25). A study done in Ethiopia showed that HAART
resulted in a decrease of mortality (adjusted HR [95%CI] = 0.35 (0.19–0.63) (32).
The use of HAART during treatment for tuberculosis significantly protected against
mortality when compared to HIV-infected patients who either did not receive any anti-
retroviral medications or received regimens other than HAART (RR 0.36, 95% C.I.
0.14-0.91, p = 0.01) (33).Not receiving ART was associated with a higher probability
of death (HR = 20.0, 95% CI = 8.62-45.45, P = 0.001)(24). In another study ART
remained strongly associated with reduced mortality during TB treatment. In
multivariate analysis of patients who did not receive ART and adjusting for CD4 count,
smear status, and hospital providing treatment, co-trimoxazole was not associated with
survival (AOR 0.9, CI 0.5–1.9) (34). The risk for death caused by TB was lower for
persons who took ART (HR 0.2, 95% confidence interval [CI], 0.1–0.5) (20).
Factors which remained independently associated with death in the multivariate model
include ART use with HR of 0.17 (95% CI: 0.12 to 0.24). Use of ﬂuconazole after TB
diagnosis remained statistically protective against death (HR 0.7, CI: 0.5 to 1.0) (26).
Having HAART started during TB treatment (HR 0.43, 95% CI 0.26–0.72) was
independently associated with long-term survival after TB diagnosis (21). ART
exposure at any time during TB treatment remained independently protective against
death (adjusted hazard ratio [HR] 0.41, 95%CI 0.28–0.60) (22).
7
3. OBJECTIVES
3.1. General objective
To assess predictors of mortality among TB-HIV co-infected Patients during TB

treatment.
3.2. Specific objectives
1. To estimate the time to death of TB-HIV co-infected Patients during TB treatment.

2. To assess the mortality rate of TB-HIV co-infected patients during TB treatment.
3. To determine independent predictors of mortality among TB-HIV co-infected
Patients during TB treatment.
8
4. METHODS AND MATERIALS
4.1. Study area and period
The study was conducted in Bahirdar town governmental health institutions, located in
northwest Ethiopia 565Km far from Addis Ababa. Felege Hiwot Refferal hospital gives
different specialized clinical services for approximately 7-9 million people in Amhara
regional state. It is a teaching Hospital for Bahirdar University with basic facilities for
HIV care and treatment and with established clinical set up and highly trained medical
personnel. Any patient diagnosed as having HIV in any of HIV counseling and testing
protocols (VCT, PICHT) and referrals from other health facilities referred to the ART
clinic for Pre-ART and ART follow up and registered in Pre-ART and ART log books
according to the status of the patients on diseases progression. The hospital and
Bahirdar health center have started pre-ART and ART services since 2005 and other
health centers have started the service in 2009. The 2003 E.C annual report of the
hospital showed that 1600 TB cases detected, 13,590 PLWHA ever enrolled in ART
clinic, 9,222 ever started ART and currently 5,547 PLWHA are taking ART.
The study was conducted from August, 2011 to January, 2011.
4.2. Study design
Institution based retrospective cohort study was used.
4.3. Population
4.3.1. Source population
All TB-HIV co-infected patients in Felege Hiwot referral hospital and three health
centers in Bahirdar town.
4.3.2. Study population
Those TB-HIV co-infected patients who started ART before TB treatment initiation or
during TB treatment were included in exposed cohort (on ART) and patients who
9
didn’t receive ART till completion of TB treatment were included in non-exposed
cohort (non-ART).
Those patients fulfilling the following inclusion and exclusion criteria at TB and ART
clinic
Inclusion criteria
 TB-HIV co-infected patients aged 15 years or older who have diagnosed TB at

any time during Pre-ART and ART follow up since April, 2009 and completed
TB treatment before January, 2012.
 Patients who had diagnosed both TB and HIV during initial visit
 Patients who had diagnosed TB during follow up care for HIV/AIDS
 TB-HIV co-infected patients with complete intake form, registers, and follow
up form from TB diagnosis to completion of TB treatment.
Exclusion Criteria include
 Transferred in and transferred out patients
4.4. Sample size determination
The sample size was calculated using two sample proportion formulas in Epi-Info
version 3.5.1 for windows. Calculation was based on the assumption that type I error of
5%, power of 80% and two exposed (ART during anti-TB treatment) to one non-
exposed (Not taking ART during anti-TB treatment).
To calculate the sample size proportion of deaths in the non-exposed cohort (24.6%)
and in the exposed cohort (11.3%) was taken from retrospective cohort study done in
India (23). The minimum sample size was 312 (208 exposed and 104 non-exposed),
adding 10% for incomplete records the final sample size was 342 (228 exposed and 114
non-exposed). But we took all eligible records from April, 2009 to January, 2012. We
reviewed 422 records of TB-HIV co-infected patients (272 exposed and 150 not
exposed).
[Let P= p1+ rp2/1+r and r =n2/n1

10
Where; P1=proportion of death among exposed
P2=proportion of death among non-exposed
α=level of significance, Zα/2=1.96 at 95% CL
Power=80%=1-β, Zβ=1.28
n=the minimum required sample size in each group]
4.5. Sampling Procedures
Felege Hiwot Referral Hospital and three health centers were selected for the study
purposely to get adequate number of sample. Profiles of all TB-HIV co-infected
patients between April 2009 and January, 2012 in TB, Pre-ART and ART registration
log book and patient’s card were reviewed. Then based on TB, Pre-ART and ART
registration log book records of TB/HIV co-infected patients were categorized into on
ART and non-ART. All records after the introduction of HMIS that fulfills the
inclusion criteria were included.
4.6. Data collection procedures
4.6.1. Variables
4.6.1.1. Independent variables
The independent variables were
 Socio-demographic characteristics including age, religion, sex, educational

level, marital status and occupational status.
 Base line and recent clinical and  cotrimoxazole
laboratory information chemoprophylaxis,
 Past opportunistic illness  WHO clinical staging
 TB type and treatment  Liver function tests (ALT/AST).
 ART treatment
11
4.6.1.2. Dependent variables
The dependent outcome was time to death from TB diagnosis to treatment completion.
4.6.2. Instruments
The data collection instruments were prepared by the principal investigator by

reviewing pre-ART, ART, HIV care follow up form and TB-registration book in ART
and TB clinic. It had included the following parts
 Socio demographic data

 Base line and recent clinical and laboratory data
 Medication and treatment
 TB location site and category
4.6.3. Data collection
Nurses who work in TB and ART clinic were selected to collect the data. Two day
training was given for all supervisors and data collectors. Pre-test was undertaken
before we collect the actual data. The data was collected by reviewing pre-ART
register, lab request, monthly cohort form, and follow up form, Anti-TB record form,
ART intake form and patients’ card. Date of death was extracted from TB registration
logbook. The most recent laboratory results during TB diagnosis and before were used
as a base line value. Supervisors monitored activities of data collectors daily and
checked completeness of each questionnaire. The overall activity was controlled by the
principal investigator of the study. All completed data collection form was examined
for completeness and consistency during data management and storage.
4.6.4. Data Quality control
Data quality was assured by Pre-tested instrument, trained data collectors and
supervisors, continuous supervision and monitoring. Data completeness and
consistency was checked before and after data entry.
12
4.7. Measurement of variables
Any death that occurred during TB treatment was classified as ‘‘death during
treatment’’. If the date of ART or CPT treatment initiation was more than 1 week
before TB treatment initiation, that person was said to be on ART or CPT prior to TB
treatment. Patients who initiated ART at any time before TB treatment was completed
were classified as having received ART during TB treatment.
Since the 2010 guideline was not implemented during the study period we used the
former definition.
Smear positive pulmonary tuberculosis (PTB+) if two or more initial sputum

examinations were positive for acid fast bacilli (AFB), or one sputum positive for AFB
plus radiographic abnormalities consistent with active TB as determined by a
physician.
Extra pulmonary tuberculosis (EPTB) refers to TB of organs other than the lungs and
diagnosis was based on strong clinical suspicion by a physician.
Functional status was measured at base line, and a person is categorized into working =
“able to perform usual work in or out of the house”; Ambulatory= “able to perform
activities of daily living” and Bedridden= “not able to perform activities of daily
living”.
4.8. Data analysis procedures
Completed data collection instruments coded using numbers. Data was entered to
EpiData 3.1(EpiData Association, Odense Denmark) for windows. SPSS version 16.0
for windows and STATA version 11.0 were used for analysis. Data was cleaned and
edited by simple frequencies and cross tabulation before analysis. Date of deaths were
recorded by reviewing TB registration logbook, medical registration in the hospital, or
registration by ART adherence supporter through calling using the registered phone
number and individuals alive at the end of TB treatment completion were censured.
Finally, the outcome of each subject was dichotomized into censored or death. Death
13
from any cause during TB treatment was listed as on-treatment TB death according to
the WHO TB treatment outcome definition (35).
The response variable was survival time, defined as time in days transpired from the
date of initial TB treatment to death or, in the case of individuals who did not die, the
time to completion of the TB treatment.
Mean (with standard deviation), median (with inter quartile range, IQR) and
frequencies (%) were used to describe patients’ characteristics in each cohort. Chi-
square test and T- test were used to compare categorical and continuous variables
between the two cohorts, respectively. Actuarial life table was used to estimate survival
after initiation of anti-TB and the Kaplan-Meier test was used to estimate the
probability of death and the median time to death after TB diagnosis. The log-rank test
was used to compare time to death between the two groups. The Cox proportional
hazard model was used to determine the chance of death after TB diagnosis by
adjusting for confounding factors. All statistically significant factors in the bivariate
analysis were included in the final model. The crude and adjusted hazard ratio (HR)
and its 95% confidence interval (CI) were estimated. A p-value less than 0.05 was
considered statistically significant.
4.9. Ethical Considerations
Ethical clearance was obtained from Review Ethics Committee (REC) of School of
Public Health, Addis Ababa University. Following the approval official letter of co-
operation was written to concerned bodies by the School of Public Health. Permission
letter was obtained from Amhara regional health bureau. As the study was conducted
through review of medical records, the individual patients was not be subjected to any
harm as far as the confidentiality was kept. Informed consent was obtained from each
participant during home visiting or calling. To preserve the confidentiality, nurses
working in ART clinic of Felege Hiwot referral hospital and three health centers
extracted the data from the medical records. Moreover, no personal identifiers were
used on data collection form. The recorded data was not accessed by a third person
except the principal investigator, and was kept confidentially.
14
5. RESULTS
5.1. Enrollment of TB-HIV co-infected patients in the study
From April 2009 – September 2011, 2,773 adult TB patients were registered for
treatment in Felege Hiwot Referral hospital, Bahirdar health center, Han health center
and Abay health center and 849 (30.6%) were known to be HIV-infected, of which 488
(57.5%) were eligible for the study [Figure 1]. We later excluded 29 patients whose
card was not found and 37 patients whose ART status was not determined. The
remaining 422 TB-HIV co-infected patients were eligible for the study. A total of 422
TB-HIV co-infected patients (272 ‘On ART’ and 150 ‘Non-ART’ cohorts) were
followed retrospectively for a median of 197 days with IQR 140-221days among on
ART and 191 days with IQR 65.5-209.5 days among non-ART cohorts. The follow up
period was slightly higher in on ART cohorts.
Adult TB patients registered from April 2009-

September 2011 (n= 2,773)
TB-HIV co-infected patients (n=849)
Not eligible (n=361)
Transfer out, n=194
Transfer in, n=57 Not enrolled, n=66

Eligible for the
Out come not registered, study n=488
ART status not
n=46
determined, n=37
HIV status known after
Patient card not
TB treatment initiation, Enrolled in the found, n=29
n=64 study
n=422
Figure 1: Enrollment of TB-HIV co-infected patients in the study, 2012
15
5.2. Baseline Socio-Demographic Characteristics of the Study Subjects
In this study both cohorts were not statistically different for each attribute of the socio-
demography. The median age of study subjects in both cohorts was 30 years with inter
quartile range (IQR) 27-37.5years and 25-38years in On ART and Non-ART cohorts
respectively. Majority of study subjects were urban resident 235(90.7%) and
129(87.2%) in On ART and Non-ART cohorts respectively. Females account for a
majority of study subjects; 141(53.4%) and 83(56.5%) in On ART and Non-ART
cohorts respectively (Table 1).
In both cohorts majority of patients were orthodox; 226(85.3%) and 117(80.2%) in On

ART and Non-ART respectively. Majority of study subjects were married in both
cohorts; 108(40.0%) and 57(39.0%) in On ART and Non-ART respectively. More than
one third of patients in both cohorts were secondary school completed; 93(34.8%) and
50(35.7%) in On ART and Non-ART respectively. More than one fourth of study
subjects in On ART cohorts were nongovernmental employees; 73(29.2%) and
32(23.4%) study subjects were government employees in Non-ART cohorts (Table 1).
16
Table 1: Baseline socio-demographic characteristics of TB-HIV co- infected patients
during TB treatment in Bahirdar town governmental health institutions, 2012
Base line variable On ART Non-ART X2 Value P-

(n=272) (n=150) (df) Value
Address(n=407)
Urban 235(90.7%) 129(87.2%) 1.271 0.260
Rural 24(9.3%) 19(12.8%) (1)
Age(n=408)
Mean ± SD 32.58±9.123 31.98±9.837 0.753♦ 0.452
Median(IQR) 30yrs(27-37.5) 30yrs(25-38)
Sex(n=411)
Male 123(46.6%) 64(43.5%) 0.355 0.551
Female 141(53.4%) 83(56.5%) (1)
Religion(n=411)
Orthodox 226(85.3%) 117(80.2%) 5.962 0.183
Muslim 26(9.8%) 18(12.3%) (2)
Others 13(4.9%) 11(7.5%)
Marital status(n=415)
Single 76(28.1%) 55(37.9%) 7.901 0.131
Married 108(40.0%) 57(39.3%) (3)
Divorced/Separated 52(19.2%) 24(16.6%)
Widowed 34(12.6%) 9(6.2%)
Educational status(n=407)
Not educated 77(28.8%) 40(28.6%) 0.765 0.858
Primary 61(22.8%) 35(25.0%) (3)
Secondary 93(34.8%) 50(35.7%)
Tertiary 36(13.5%) 15(10.7%)
Occupational Status
House wife 18(7.2%) 5(3.6%) 7.932 0.239
Merchant 39(15.6%) 16(11.7%) (6)
Government employee 46(18.4%) 27(19.7%)
Nongovernment employee 73(29.2%) 32(23.4%)
Day laborer 43(17.2%) 31(22.6%)
Jobless 19(7.6%) 16(11.7%)
Others 1,2,3 12(4.8%) 10(7.3%)
1. Student 2. Farmer 3.Commercial sex worker, ♦=T-test Statistic for independent sample test used
17
5.3. Clinical and immunological characteristics of the Study Subjects
When we see clinical condition of study subjects, there was no statistical difference in
both cohorts in most variables except for history of prophylactic medication in which
higher proportion (51.6%) of patients on ART had history of prophylactic medication
compared to patients not on ART (27.1%) (X2=21.721; df (1); p=0.000). Among the
study subjects more than one third had at least one past opportunistic infection;
105(43.6%) and 55(44.0%) in On ART and Non-ART cohorts respectively. And also
58(22.4%) and 31(21.7) study subjects had history of past TB treatment in On ART and
Non-ART cohorts respectively (Table 2).
Majority of study subjects were able to perform their usual work (working);
159(60.9%) and 80(58.8%) in On ART and Non-ART cohorts respectively. Majority of
study subjects were in WHO clinical stage III during initiation of TB treatment;
181(67.3%) and 94(62.7%) in On ART and Non-ART cohorts respectively (Table 2).
The result showed that On ART and Non-ART cohorts were statistically different in
CD4 count (T=10.305; p=0.000) and liver function tests (T=2.081; p=0.039 and
T=2.904; 0.004 for ALT and AST respectively). But both cohorts were not different in
hemoglobin level. Study subjects in On ART cohort had much lower CD4+ cell count
with median of 114cells/µl and IQR 58-185cells/ µl than Non- ART with median of
291cells/µl and IQR 183.5-448cells/µl. In this study hemoglobin level is not
statistically different in both cohorts (T=0.063; p=0.950) with the median of 11.3mg/dl
(IQR 10.0-13.0) and 12.0 mg/dl (IQR 10.1-13.0) in On ART and Non-ART cohorts
respectively (Table 2).
18
Table 2: Baseline clinical and immunological characteristics of TB-HIV co- infected
patients during TB treatment in Bahirdar town governmental health institutions, 2012
Base line variable On ART Non-ART X2 Value P-

(n=272) (n=150) (df) Value
Past OIs(n=366)
Yes 105(43.6%) 55(44.0%) 0.006 0.937
No 136(56.4%) 70(56.0%) (1)
Past TB Treatment(n=402)
Yes 58(22.4%) 31(21.7%) 0.027 0.869
No 201(77.6%) 112(78.3%) (1)
Past Prophylactic medication(n=387)
Yes 131(51.6%) 36(27.1%) 21.372 0.000*
No 123(48.4%) 97(72.9%) (1)
Functional status (n=397)
Working 159(60.9%) 80(58.8%) 1.363 0.506
Ambulatory 72(27.6%) 44(32.4%) (2)
Bedridden 30(11.5%) 12(8.8%)
WHO Staging(n=419)
Stage III 181(67.3%) 94(62.7%) 0.911 0.34
Stage IV 88(32.7%) 56(37.3%) (1)
CD4 count(cells/µl) (n=408)
Mean ± SD 132.9±94.42 312.78±192.8 10.305♦ 0.000*
Median(IQR) 114(58-185) 291(183.5-448)
Hgb level (mmHg)(n=357)
Mean ± SD 11.36±2.3 11.46±1.83 0.063♦ 0.95
Median(IQR) 11.3(10.0-13.0) 12.0(10.12-13.0)
ALT(n=336)
Mean ± SD 35.6±20.86 42.73±25.84 2.081♦ 0.039*
Median(IQR) 29(22.0-40.025) 34(51.76-61.25)
AST(n=303)
Mean ± SD 42.05±23.11 51.76±29.27 2.904♦ 0.004*
Median(IQR) 36(27.0-54.0) 47(26.0-69.0)
* Significant at α=0.05, ♦=T-test Statistic for independent sample test used
19
5.4. Characteristics of study subjects on type of TB diagnosis, type and medication
There was great difference to statistically significant level in type of TB diagnosis

between the cohorts. Majority of study subjects in On ART group had smear negative
PTB; 107(39.3%) whereas only 36(24.0%) had smear negative PTB in Non-ART group
(X2 =10.434; df=2; p=0.005) (Table 3).
In both cohorts majority of study subjects had new form of TB; 258(94.9%) and 146
(97.6%) in On ART and Non-ART groups respectively. More than three fourth of study
subjects had taken 2HRZE/6EH anti-TB regimen; 214(78.7%) and 119(79.3%) in On
ART and Non-ART cohorts. Higher proportion (93.3%) of study subjects in On ART
cohort had got cotrimoxazole prophylactic therapy compared to Non-ART cohorts
(77.0%) (Table3).
5.5. Base line social condition
There was statistical difference in the groups regarding to religious/supportive care

(X2=7.977; df=1; p=0.005). Among On ART group only 49(20.4%) had
religious/supportive care whereas 43(33.9%) of study subjects in Non-ART group had
religious/supportive care. Majority of study subjects in both cohorts had disclosed their
HIV sero status; 209(83.6%) and 97(77.0%) in On ART and Non-ART groups
respectively. Similarly; at least one general concern was identified in majority of study
subjects; 164(69.2%) and 93(72.7%) in On ART and Non-ART cohorts respectively.
(Table 4)
20
Table 3: Type of TB diagnosis and medication of TB-HIV co-infected patients during
TB treatment in Bahirdar town governmental health institutions, 2012
Variable On ART Non-ART X2 Value P-

(n=272) (n=150) (df) Value
TB diagnosis
Smear Positive PTB 53(19.5%) 40(26.7%) 10.434 0.005*
Smear Negative PTB 107(39.3%) 36(24.0%) (2)
Extra PTB 112(41.2%) 74(49.3%)
TB Type
New 258(94.9%) 146(97.3%) 1.457 0.227
Not new 14(5.1%) 4(2.7%) (1)
Anti-TB Regimen
2HRZE/6EH 214(78.7%) 119(79.3%) 12.587 0.002*
2HRZE/4RH 26(9.6%) 26(17.3%) (2)
Other 32(11.7%) 5(3.3%)
CPT (n=397)
Prescribed 239(93.0%) 109(77.9%) 19.19 0.000*
Not Prescribed 18(7.0%) 31(22.1%)
Outcome of TB Treatment(n=417)
Cure 37(13.7%) 16(11.0%) 8.039 0.045*
Treatment completed 167(61.6%) 77(52.7%) (3)
Defaults 18(6.6%) 9(6.2%)
Death 49(18.1%) 44(30.1%)
* Significant at α= 0.05
21
Table 4: Baseline social condition of TB-HIV co- infected patients during TB
treatment in Bahirdar town governmental health institutions, 2012
Variable On ART Non-ART X2 Value P-Value

(n=272) (n=150) (df)
Religious/supportive care(n=367)
Yes 49(20.4%) 43(33.9%) 7.977 0.005*
No 191(79.6%) 84(66.1%) (1)
HIV Serostatus disclosure(n=376)
Yes 209(83.6%) 97(77.0%) 2.424 0.120
No 41(16.4%) 29(23.0%) (1)
General concern Identified(n=365)
Yes 164(69.2%) 93(72.7%) 0.479 0.490
No 73(30.8%) 35(27.3%) (1)
*Significant at α= 0.05
5.6. Comparison of mortality between On ART and Non-ART cohorts
All 422 study subjects had contributed a total of 2274.4person months; 1545.03person
months in on ART cohort and 729.37person months in non-ART cohort respectively. In
Non-ART cohort more TB-HIV co-infected patients died during TB treatment;
44(29.3%) than On ART cohort; 49(18.0%) with incidence rate of 6.03Per 100Person
month observations (95%CI: 4.5, 8.1) and 3.20 per 100PMO (95% CI: 2.40, 4.20). The
overall incidence rate of mortality during TB treatment is 4.09 per 100person month
observations (95% CI: 3.34, 5.01). The result showed that incidence of mortality in the
first month of TB treatment is 5.40 per 100PMO and 4.8 per 100PMO in second month
of TB treatment in On ART cohort. The corresponding value in Non-ART cohort is
16.90 per 100PMO and 5.90 per 100PMO in the first and second month of TB
treatment respectively.
The median time for development of mortality during TB treatment has shown
variation between the two groups. The median time to develop death was 29.5 days and
59 days in On ART cohort and non-ART cohort respectively. In both cohorts majority
of deaths occur in the first month of TB treatment (0-28 days) and decreases in later
follow up period. Based on actuarial life table analysis the probability of not
developing death at 28 days and 56 days in On ART cohort were 95% and 91%
22
respectively. The corresponding values in Non-ART cohort were 86% and 80%
respectively (Table 5).
Table 5: Actuarial life table analysis of TB-HIV co- infected patients during TB
treatment in Bahirdar town governmental health institutions, 2012
ART Time Number Number Number Number Probability Cumulative

status interval entering with exposed of of not Probability
in days interval drawing to risk deaths developing of not
during death developing
interval death
0-28 272 5 269.5 13 0.95 0.95
28-56 254 0 254 10 0.96 0.91
56-84 244 4 242 7 0.97 0.89
On ART 84-112 233 2 232 6 0.97 0.86
112-140 225 13 218.5 7 0.97 0.84
140-168 205 24 193 3 0.98 0.82
168-196 178 33 161.5 1 0.99 0.82
196-224 144 84 102 1 0.99 0.81
224-252 59 57 30.5 1 0.97 0.78
0-28 150 5 147.5 20 0.86 0.86
28-56 125 0 125 10 0.92 0.80
56-84 115 0 115 5 0.96 0.76
Non- 84-112 110 2 109 3 0.97 0.74
ART
112-140 105 11 99.5 2 0.98 0.72
140-168 92 10 87 1 0.99 0.72
168-196 81 17 72.5 1 0.99 0.71
196-224 63 40 43 1 0.98 0.69
224-252 22 21 11.5 1 0.91 0.63
23
The Kaplan-Meier analysis and the log-rank test were used to compare survival
probabilities of the two groups. Overall probability of survival in On ART cohort was
significantly greater than Non-ART cohort, i.e. the risk of developing death is higher in
Non-ART cohort (log rank statistic=8.93, df=1, P=0.003) (Figure 2, 3).
Figure 2: Kaplan-Meier estimate of survival among TB-HIV co-infected patients

institution, 2012

institution, 2012
After stratification of the cohorts based on type of TB diagnosis, the result showed that
On ART cohort had significantly greater chance of not developing death during TB
treatment than Non-ART cohort across strata smear positive and smear negative PTB
(P=0.030 and P=0.003 respectively), but significant difference was not observed in the
stratum of extra PTB (P=0.732). Furthermore; stratification by CD4 count showed that
24
On ART cohort had significantly greater chance of not developing death during TB
treatment than Non-ART cohort across strata CD4 level 0-75cells/µL and CD4 level
75-150cells/ µL (p=0.000 and 0.014 respectively). There was no significant difference
in both cohorts in developing death during TB treatment in CD4 level 150-250cells/µl,
and >=250cells/µl stratums (p=0.233 and p=0.593 respectively) (figure 3-6).

TB treatment with and without ART among smear positive PTB patients in Bahirdar
town governmental health institutions, 2012

TB treatment with and without ART among smear negative PTB patients in Bahirdar
town governmental health institutions, 2012
25
during TB treatment with and without ART among extra PTB patients in Bahirdar town
governmental health institutions, 2012

during TB treatment with and without ART among CD4 level 0-75cells/µL patients in
Bahirdar town governmental health institutions, 2012
26
5.7. Predictors of mortality in TB-HIV co-infected patients during TB treatment
Bivariate Cox-proportional model was used to assess the relationship between variables
and the risk of developing mortality. Before fitting the covariate into the model
proportional hazard assumption was checked by examining log (-log) S (t) plots. The
result showed that health institution type, educational status, functional status, CD4
level, hemoglobin level, ALT level, AST level, type of TB diagnosis, ART treatment
and CPT prophylaxis were significantly associated with developing mortality in TB-
HIV co-infected patients during TB treatment (Table 6-8).
The bivariate analysis showed that the risk of developing death in patients treated TB at
hospital is 2.18 times (95% CI: 1.43-3.33) higher compared to TB patients at health
center. Compared to the reference group; TB patients in the age category 45years and
above (HR=2.58, 95% CI: 1.34-4.92), ambulatory and bedridden functional status
(HR=2.76, 95% CI: 1.71-4.47 and HR=3.88, 95% CI: 2.15-7.02 respectively) and CD4
count less than 75cells/µl (HR=2.08, 95% CI: 1.17- 3.70) increased risk of mortality
during TB treatment. In the study, completing primary education reduced risk of death
by 55% (HR=0.45, 95% CI: 0.22-0.90) compared to not educated TB-HIV co-infected
patients. In addition the risk of mortality decreased for every 1mmHg increase in
hemoglobin by 9.5%.
In this study the risk of death increased by 1.4% for 1unit increase in ALT and AST
(p=0.001 and p=0.000) respectively. The result showed that compared to smear
negative PTB patients; smear positive PTB patients and extra PTB patients had 2.02
(95% CI: 1.07-3.83) and 2.77 (95% CI: 1.61-4.75) times higher risk of death
respectively. In addition patients who didn’t start CPT were 3.15 times higher at risk of
developing death (95% CI: 1.95-5.11).
ART is another important predictor of mortality. The result showed that the risk of
death decreased by 46.0% (HR=0.54, 95% CI: 0.36-0.82) in TB-HIV co-infected
patients who have started ART. In the bivariate analysis sex, religion, past history of
OIs, past TB treatment, weight and WHO staging were not associated with risk of death
during TB treatment.
27
Table 6: Medication and TB location associated predictors of mortality among TB-
HIV co-infected patients during TB treatment in Bahirdar town governmental health
institutions, 2012
Variable Number Number Incidence of Mortality Crude Hazard

at risk of death per 100 person month Ratio (95% CI)
observation (95% CI)
ART
Started 272 49 3.17 (2.39, 4.19) 0.54 (0.36, 0.82)*
Not started 150 44 6.03 (4.49, 8.11) 1
CPT prophylaxis
Prescribed 348 60 3.59 (2.66, 4.87) 1
Not prescribed 49 23 10.18 (6.77, 15.32) 3.15 (1.95, 5.11)*
TB Location Site
Smear negative PTB 143 17 2.02 (1.26, 3.25) 1
Smear positive PTB 93 21 4.31 (2.81, 6.61) 2.02 (1.07, 3.83)*
Extra PTB 186 55 5.81 (4.46, 7.57) 2.77 (1.61, 4.78)*
28
Table 7: Clinical and immunological associated predictors of mortality among TB-HIV
institutions, 2012
Variable Number Number Incidence of Crude Hazard

at risk of death mortality per 100 Ratio (95% CI)
person month
observation
Past OIs
No 206 50 4.62 (3.50, 6.09) 1
Yes 160 31 3.53 (2.48, 5.02) 0.77 (0.49, 1.20)
Past TB treatment
No 313 66 3.95 (3.10, 5.03) 1
Yes 89 18 3.58 (2.25, 5.68) 0.91 (0.54, 1.54)
Functional status
Working 239 31 2.21 (1.55, 3.14) 1
Ambulatory 116 36 6.39 (4.61, 8.87) 2.77 (1.71, 4.47)*
Bedridden 42 17 7.31 (3.81, 14.05) 3.88 (2.15, 7.02)*
WHO Staging
Stage III 275 56 3.69 (2.84, 4.80) 1
Stage IV 144 36 4.86 (3.51, 6.74) 1.30 (0.86,1.98)
CD4 count
< 75 107 33 6.22 (4.42, 8.75) 2.08 (1.17, 3.30)*
75-150 91 21 4.44 (2.90, 6.81) 1.50 (0.80, 2.82)
150-250 95 19 3.57 (2.28, 5.59) 1.24(0.65, 2.37)
>=250 115 18 2.89 (1.82, 4.59) 1
Hgb count 357 88 4.66 (3.69, 5.63) 0.90 (0.82, 0.99)*
ALT 336 79 4.41 (3.44, 5.38) 1.01 (1.01, 1.02)*
AST 303 71 4.35 (3.34, 5.36) 1.01 (1.01, 1.02)*
29
Table 8: Socio-demography associated predictors of mortality among TB-HIV co-
infected patients during TB treatment in Bahirdar town governmental health
institutions, 2012
Variable Number Number Incidence of Crude Hazard

at risk of death Mortality per 100 Ratio (95% CI)
person month
observation (95% CI)
Health Institution
Health center 226 34 2.67 (1.91, 3.72) 1
Hospital 196 59 5.89(4.57, 7.61) 2.18 (1.43, 3.33)*
Age(n=408)
15-24 100 18 3.30 (2.08, 5.24) 1
25-34 193 40 3.79 (2.78, 5.17) 1.15 (0.66, 2.01)
35-44 69 16 4.40(2.70, 7.18) 1.30 (0.66, 2.55)
>=45 46 19 8.97 (5.72, 14.06) 2.58 (1.34, 4.92)*
Sex
Male 187 45 4.53 (3.36, 6.07) 1
Female 224 45 3.69(2.75, 4.94) 0.82 (0.54, 1.24)
Religion
Orthodox 343 66 3.52 (2.76, 4.48) 1
Muslim 44 13 5.53 (3.21, 9.52) 1.58 (0.87, 2.87)
Others 24 8 6.81 (3.40, 13.61) 1.93 (0.93, 4.02)
Educational status
Not educated 117 26 4.47 (3.05, 6.57) 1
Primary 96 11 1.93 (1.07, 3.49) 0.45 (0.22, 0.90)*
Secondary 143 37 4.82 (3.49, 6.66) 1.10 (0.66, 1.81)
Tertiary 51 13 4.48 (2.60, 7.72) 1.05 (0.54, 2.05)
To identify independent predictors of mortality during TB treatment, a multivariate

Cox-Proportional hazard adjusted model was fitted with the variables having a P-value
30
< 0.05 in the bivariate analysis. Before running the final model multicollinearity was
checked using variance inflation factor (VIF). When hemoglobin level and liver
function tests (ALT and AST) were included in the model, numbers of valid cases were
inadequate to have appropriate power of the study. These variables were excluded from
multivariate analysis even if they were statistically significant in the bivariate analysis.
Accordingly, receiving ART, CPT, CD4 count, functional status, type of TB diagnosis
and health institution type were independent predictors of mortality after controlling for
the other factors. From these factors; receiving ART during TB treatment had an
independent protective benefit against risk of mortality and decreased risk of mortality
by 65% (AHR=0.35, 95% CI: 0.19-0.64). In addition CPT remained important factor in
reduction of mortality during TB treatment in which patients without CPT were 3.03
times higher at risk of mortality (95% CI: 1.58, 5.79). CD4 count category 0-75cells/µl,
75-150cells/µl and 150-250cells/µl, extra PTB type, being ambulatory, and treatment in
hospital were independent predictors of increased risk of mortality during TB treatment
(Table 9).
31
Table 9: Multivariate analysis of variables associated with mortality among TB-HIV
institutions, 2012
Variable Crude Hazard Adjusted Hazard

Ratio (95% CI) Ratio (95%) CI )
ART
Started 0.54 (0.36, 0.82) 0.35 (0.19, 0.64)*
Not started 1 1
CPT prophylaxis
Prescribed 1 1
Not prescribed 3.15 (1.95, 5.11) 3.03 (1.58, 5.79)*
TB diagnosis
Smear negative PTB 1 1
Smear positive PTB 2.02 (1.07, 3.83) 2.11 (0.95, 4.65)
Extra PTB 2.77 (1.61, 4.78) 2.39 (1.23, 4.66)*
CD4 count
< 75 2.08 (1.17, 3.30) 4.83 (1.98, 11.78)*
75-150 1.50 (0.80, 2.82) 3.57 (1.48, 8.61)*
150-250 1.24(0.65, 2.37) 3.07 (1.33, 7.07)*
>=250 1
Functional status
Working 1 1
Ambulatory 2.77 (1.71, 4.47) 2.10 (1.22, 3.62)*
Bedridden 3.88 (2.15, 7.02) 2.11 (0.98, 4.53)
Health Institution
Health center 1 1
Hospital 2.18 (1.43, 3.33) 2.64 (1.51, 4.62)*
Age(n=408)
15-24 1 1
25-34 1.15 (0.66, 2.01) 1.17 (0.60, 2.29)
35-44 1.30 (0.66, 2.55) 0.98 (0.43, 2.23)
>=45 2.58 (1.34, 4.92) 2.20 (0.97, 4.59)
Educational status
Not educated 1 1
Primary 0.45 (0.22, 0.90) 0.49 (0.22, 1.12)
Secondary 1.10 (0.66, 1.81) 0.88 (0.49, 1.58)
Tertiary 1.05 (0.54, 2.05) 0.65 (0.29, 1.47)
*Significant at α= 0.05. ART, health institution, age, educational status, functional status, CD4 count, type
of TB diagnosis and CPT prophylaxis were included in the model.
32
6. DISCUSSION
This study provides information on the overwhelming problem of the high mortality of
TB-HIV co-infected patients during TB treatment. In this resource limited setting, more
than 1 in 5 TB-HIV co-infected individuals died during TB treatment. The results from
this study demonstrated that ART remained independently protective against mortality
during TB treatment.
In this study, the median age of TB-HIV co-infected patients was 30 years consistent
with other study (34). This is slightly lower compared to similar studies in developing
countries (23, 25, 26, 36) and in the age group 25-44 years contributed around two third
of study subjects in both groups. Like other developing countries, the most affected
population group by HIV is in the most reproductive age group. Different studies
revealed that active TB is most common in HIV positive patients between 25 to 44
years of age.
More than half (54.5%) of the study subjects are females. This result is inconsistent
with other studies in developing countries (23, 24, 26, 36). A study done in India
showed that 67% were males (23) and another study in Thailand showed that only 34%
of study subjects were females (26). This difference may be due to high prevalence of
HIV among females.
It was difficult to get the exact value of CD4 at diagnosis of TB. However, it was tried
to look CD4 count of patients during TB treatment in the course of their follow up. In
our study the median CD4 count was more than twice in Non-ART cohort compared to
On ART cohort (291cells/µl (IQR: 183.5-448) and 114cells/µl (IQR: 58-185))
respectively). These groups of patients may be diagnosed as having HIV and TB earlier
before their clinical and immunological condition deteriorate. The median CD4 count
in this study is much higher compared to other studies (20, 21, 23-26, 36, 37). The
difference may be due to in this study CD4 count was taken during TB treatment or one
month before TB treatment initiation and most study subjects were started ART before
TB diagnosis which may improve their immunological status. In addition the result
showed that 86.3% of study subjects had CD4 count below 350cells/µl indicating
33
progressive immunodeficiency. This is similar with a study done in Zimbabwe 84.6%
had CD4 count below 350cells/µl (38).
Although there was no statistica1 difference on type of TB diagnosis in cohorts, 55.9%

and 44.1% of study subjects diagnosed with PTB and extra PTB respectively. This is in
line with other studies (24, 26, 30) but proportion of extra PTB in this study is high
compared to two studies done in India (22.9% and 31%) and study done in Thailand
reported 31% of study subjects had extra PTB (23, 25, 34). The variation may be due to
difference in TB location site classification and diagnosis of TB.
We found that mortality rate was high (22%) among TB-HIV co-infected patients
during TB treatment. In line to this, previous study have reported high mortality among
TB-HIV co-infected patients prior to successful completion of TB treatment ranges
from 8.5% to 30% (4, 20-23, 25, 26, 30, 34, 37). In this study, death occurred in 49 of
272 (18.0%) patients exposed to ART during TB treatment compared with 44 of 150
(29.3%) patients without ART. The finding is similar with a study reported in India
where death occurred in 11.3% and 24.6% patients respectively (23). In this study the
proportion of death in patients who didn’t start ART was lower (29.3%) compared with
46% reported from a study done in Thailand (26). Another study in Thailand also
reported that 5 (7%) of 71 patients who received ART died compared with 94 (43%) of
219 patients who did not receive ART (RR 0.2; 95% CI: 0.1–0.4) (34). In Malawi a
total of 132 (20%) patients died during the 8-month course of anti-tuberculosis
treatment which is consistent with our finding (22%) (39).
In this study, we have documented risk of mortality was high in the first month of TB
treatment with incidence rate of 9.3 per 100PMO (95% CI: 6.8-12.9). This may be due
to delayed presentation of patients and thus advanced TB and HIV/AIDS, late diagnosis
of TB within health institutions and life-threatening HIV-related complications. This is
similar with a study done in Thailand (27). Another study conducted in Sub-Saharan
Africa also documented the rationale for starting ART soon after TB diagnosis is that
case-fatality among HIV-TB patients occurs mainly in the first 2 months of TB
treatment (40).
34
This study showed that nearly two third (64.5%) of TB-HIV co-infected patients had
access to ART during TB treatment. This is much higher compared to studies in
Thailand 41% and India 52% TB-HIV patients had access to ART (23, 27). The
difference may be due to emphasis of recommendations to initiate ART during TB
treatment and their study was done before some years ago which may not be the case at
this time.
In this retrospective institution based study we found that TB-HIV co-infected patients
provided ART during TB treatment had low risk of death as those not provided
ART(AHR=0.35, 95% CI: 0.19-0.64). This is consistent with other several settings (20-
24, 26, 30, 33, 34, 37, 41) in which plenty of studies precisely demonstrate the impact
of ART on the survival outcomes among TB-HIV co-infected patients with successful
immune restoration and reductions in morbidity and mortality.
International guidelines recommend that HIV-infected persons initiate treatment with

nevirapine, zidovudine, and lamivudine; because rifampin can alter drug levels of
nevirapine, guidelines recommend that efavirenz replace nevirapine in patients
receiving rifampin during TB treatment (19, 42). In line with this our finding showed
that 70.2% of patients have got efavirenze based ART regimens during TB treatment.
This may have effect on mortality by reducing risk of adverse reactions.
In addition to ART, we found other immunological factors associated with death.

Mortality rates increased in patients with lower CD4 category. In this study, TB-HIV
co-infected patients with CD4 count category 0-75cells/µl, 75-150cells/µl and 150-
250cells/µl had increased risk of mortality (AHR=4.83, 95%CI: 1.98-11.77,
AHR=3.57, 95% CI: 1.48-8.61 and AHR=3.07, 95%CI: 1.33, 7.07 respectively). This
finding is consistent with other studies in Thailand (26, 34) and in Zimbabwe CD4
count of <50 cells/μl associated with a 13-fold increased risk of death (38). In contrary,
study done in south India showed that lower CD4 count of 200/mm3 indicating
advanced immunosuppressant was not associated with mortality (30). The difference
may be in our study CD4 count categorized in small interval which can enable to see
effect of CD4 count on mortality.
35
The risk of death during TB treatment was higher in patients treated at hospital
compared to health center. But in other study hospital providing treatment strongly
associated with reduced mortality during TB treatment (34). We didn’t conclude that
hospitalization increases mortality but the possible reason may be patients who get
worse and clinically deteriorated during TB diagnosis admitted to hospital and those
who are not severely ill transferred to health centers for TB treatment. Patients who
were ambulatory had increased risk of developing death compared to working patients.
This may be due to being seriously sick secondary to deteriorated clinical and
immunological condition. Unfortunately bedridden patients didn’t show association in
multivariate analysis. This might be due to small proportion of bedridden TB-HIV co-
infected patients.
In this study TB-HIV co-infected patients with extra PTB were at increased risk of
mortality during TB treatment compared to smear negative PTB patients. WHO also
documented that case-fatality is higher in people living with HIV with smear-negative
pulmonary and extra pulmonary TB, as these patients are generally more immune-
suppressed than those with smear-positive TB (43). Our finding is not consistent with
other studies (26, 34). In other studies PTB is associated with high risk of mortality
(23). The possible reason may be HIV infected patients with extra PTB were highly
immune-compromised.
In one study conducted in Thailand risk of death during TB treatment increases as the
age of patient increase (26). But in our study age was not associated with mortality
during TB treatment. This might be due to majority of TB-HIV co-infected patients in
this study were in the younger age group.
Not initiation of CPT was associated with high risk of mortality. In line to this, studies
in district of South India and Sub-Saharan Africa showed that CPT was significantly
associated with mortality (30, 40). But in our study likelihood of not taking CPT may
be due to early deaths after diagnosis of TB and HIV. This may over estimate benefit of
cotrimoxazole prophylactic therapy.
36
7. STRENGTH AND LIMITATIONS
Strength
This study considers time of event for analysis which enables us to consider
contribution of censored study subjects.
Limitations
Our study was subjected to several important limitations.
Information about other biomedical predictors for death that may have confounded this
study, such as drug resistance, severity of immune suppression, or co morbidities, were
not available.
We were also unable to collect adequate information about specific type of extra
pulmonary TB, patient recent CD4 count.
Data about adverse events and causes of death, which are critical to assessing the risks
of combined ART and TB treatment, were not collected.
Since most deaths occur at home, it was difficult to trace all deaths.
Exclusion of patients who transferred out may bias our results slightly.
37
8. CONCLUSIONS
Significant difference was observed in mortality rate during TB treatment between on

ART and non-ART cohorts. Incidence rate of mortality during TB treatment is
significantly higher in ART not started TB-HIV co-infected patients. In both On ART
and Non-ART TB-HIV co-infected patients mortality was high in the first month of TB
treatment.
Despite the availability ART from health institutions, death was common and was
strongly associated with the absence of ART during TB treatment.
Risk of death was over 65% lower in TB-HIV co-infected patients treated with ART
compared with those not treated with ART.
In multivariate analysis being ambulatory, CD4 count category 0-75cells/µl, 75-

150cells/µl and 150-250cells/µl, extra pulmonary TB, not initiation of CPT and
treatment in hospital were independent predictors of increased risk of mortality during
TB treatment.
38
9. RECOMMENDATIONS
The country should move to start ART for all TB-HIV co‐infected patients
irrespective of CD4 levels as per the WHO recommendation.
Expanding use of ART is critical to improve the survival of TB-HIV co-infected

patients.
Reducing mortality in this population will require intensive collaboration between HIV
and TB programs and should target high levels of ART uptake and closely monitor
progress in implementation.
Further prospective studies should be conducted to identify risk of mortality during TB

treatment.
39
10. REFERENCES
1. Implementation Guideline for TB/HIV Collaborative Activities in Ethiopia. Addis

Ababa, Ethiopia: FMOH; 2007.
2. Interim policy on collaborative TB/HIV activities. Geneva: World Health

Organization; 2004.
3. Kwan CK, Ernst JD. HIV and Tuberculosis: a Deadly Human Syndemic Clin.
Microbiol. Rev Apr. 2011; 24(2):351-76.
4. Dean GL, Edwards SG, Ives NJ, et al. Treatment of tuberculosis in HIV-infected
persons in the era of highly active antiretroviral therapy. AIDS2002; 16:75-83.
5. Harries A, Maher D, Graham S. TB/HIV: a clinical manual 2 nd ed. Geneva,

Switzerland: World Health Organization; 2004.
6. Treatment of tuberculosis: guidelines 4th ed. Geneva, Switzerland: World Health

Organization; 2010.
7. Burman WJ. Issues in the management of HIV-related tuberculosis. Clin Chest

Med 2005; 2:283-94.12.
8. TB/HIV research priorities in resource-limited settings. Geneva, Switzerland:

World Health Organization; 2010.
9. Rana FS, Hawken MP, Mwachari C, et al. Autopsy study of HIV-1positive and
HIV-1-negative adult medical patients in Nairobi, Kenya. J Acquir Immune Defic
Sundr2000; 24:23-9.
10. Global Report on AIDS Epidemic. UNAIDS; 2010.
11. Global Tuberculosis control Report. Geneva, Switzerland: World Health

Organization; 2011.
12. Afework K, Mengistu G, Ayele B. Co-infection and clinical manifestation of TB in

HIV infected and uninfected adults at teaching hospital, northwest Ethiopia. J
Microbial Immunol Infect 2007; 40:116-2.
13. Maher D, Anthony H, Getahun H. Tuberculosis and HIV interaction in sub-

Saharan Africa: impact on patients and programmes;implications for policies.
Tropical Medicine and International Health august 2005; 10(8):734-42.
40
14. Single Point HIV Prevalence Estimate. Addis Ababa, Ethiopia: Federal
Democratic Republic of Ethiopia Ministry of Health (FMOH), HIV Prevention and
Control Office; 2007.
15. Global tuberculosis control: a short update to the 2010 report. Geneva, World
Health Organization; 2010 (WHO/HTM/TB/20107).
16. Report on progress towards implementation of the UN Declaration of Commitment

on HIV/AIDS. Federal Democratic Republic of Ethiopia Federal HIV/AIDS
Prevention and Control Office; 2010.
17. Deribew A, Tesfaye M, Hailmichael Y, Negussu N, Daba S, Wogi A, et al.

Tuberculosis and HIV co-infection: its impact on quality of life. Health and
Quality of Life Outcomes 2009; 7:105.
18. Scaling up Antiretroviral Therapy in Resource-limited Settings. Guidelines for a

Public Health Approach. Geneva, Switherland: World Health Organization; 2003.
19. Anti-retroviral therapy for HIV infection in adults and adolescents:

Recommendations for a public health approach. Geneva, Switzerland: World
Health Organization; 2006.
20. Cain KP, Anekthananon T, Burapat C, Akksilp S, Mankhatitham W, Srinak C, et

al. Causes of Death in HIV-infected Persons Who HaveTuberculosis, Thailand.
Emerging Infectious Diseases February 2009; 15(2).
21. Gadkowski LB, Hamilton CD, Allen M, Fortenberry ER, Luffman J, Zeringue E,
et al. HIV-Specific Health Care Utilization and Mortality among Tuberculosis/HIV
Co-infected Persons. AIDS Patient Care STDS October 2009; 23(10):845-51.
22. Straetemans M, Glaziou P, Bierrenbach AL, Si Smanidis C, Van der Werf MJ.
Assessing Tuberculosis Case Fatality Ratio: A Meta-Analysis. PLoSONE2011;
6(6):e20755J.
23. Raizada N, Chauhan LS, Babu BS, Thakur R, Khera A, et al. Linking HIV-
Infected TB Patients to Cotrimoxazole Prophylaxis and AntiretroviralTreatment in
India. PLoS ONE2009; 4(6):e5999.
24. Manosuthi W, Chottanapand S, Thongyen S, Chaovavanich A, Sungkanuparph S.

Survival Rate and Risk Factors of Mortality Among HIV/Tuberculosis-Coinfected
41
Patients With and Without Antiretroviral Therapy. Acquir Immune Defic Syndr
September 2006; 43(1).
25. Varma JK, Nateniyom S, Akksilp S, Mankatittham W, Sirinak C,

Sattayawuthipong W, et al. HIV care and treatment factors associated with
improved survival during TB treatment in Thailand: an observational study. BMC
Infectious Diseases April 2009; 42(9).
26. Sanguanwongse N, Cain KP, Suriya P, Nateniyom S, Yamada N,

Wattanaamornkiat W, et al. Antiretroviral Therapy for HIV-Infected Tuberculosis
Patients Saves Lives but Needs to Be Used More Frequently in Thailand. J Acquir
Immune Deﬁc Syndr 2008; 48:181-9.
27. Moolphate S, Aung MN, Nampaisan O, Nedsuwan S, Kantipong P, Suriyon N, et

al. Time of highest tuberculosis death risk and associated factors: an observation
of 12 years in northern Thailand. International Journal of General Medicine
February 2011; 4:181-90.
28. Pereira Domingos M, Teixeira W, Enrico C, Colosimo A. Mortality, TB/HIV co-

infection, and treatment dropout: predictors of tuberculosis prognosis in Recife,
Pernambuco State, Brazil. Cad Saúde Pública, Rio de Janeiro 2008; 24(4):887-96.
29. Mulissa Z, Jerene D, B L. Patients Present Earlier and Survival Has Improved, but
Pre-ART Attrition Is High in a Six-Year HIV Cohort Data from Ethiopia. PLoS
ONE 2010; 5:10 e13268.
30. Vijay S, Kumar P, Chauhan LS, Narayan Rao SV, Vaidyanathan P. Treatment
Outcome and Mortality at One and Half Year Follow-Up of HIV Infected TB
Patients Under TB Control Programme in a District of South India. PLoS
ONE2011; 6(7):e21008.
31. Jerene D, Endale A, Yewubnesh H, Lindtjørn B. Antiretroviral therapy at a district

hospital in Ethiopia prevents death and tuberculosis in a cohort of HIV patients.
BMC Infectious Diseases2006; 6:136.
32. Jerene D, Næss A, Lindtjørn B. Antiretroviral therapy at a district hospital in

Ethiopia prevents death and tuberculosis in a cohort of HIV patients. AIDS
Research and Therapy 2006; 3:10.
42
33. Nahid P, Gonzalez LC, Rudoy I, Jong BCd, Unger A, Kawamura M, et al.
Treatment Outcomes of Patients with HIV and Tuberculosis. Am J Respir Crit
Care Med 2003; 167:603.
34. Akksilp S, Karnkawinpong O, Wattanaamornkiat W, Viriyakitja D, Monkongdee

P, Sitti W, et al. Antiretroviral therapy during tuberculosis treatment and marked
reduction in death rate of HIV infected patients, Thailand. Emerg Infect Dis2007;
13(7):1001-7
35. Treatment of Tuberculosis: Guidelines for National Programmes 3rd ed. Geneva,
Switzerland: World Health Organization; 2003.
36. Umphonsathien M, Sungkanuparph S. Early Initiation of Antiretroviral Therapy in

HIV/Tuberculosis Co-infection and Immune Reconstitution Inflammatory
Syndrome. J Infect Dis Antimicrob Agents 2011; 28:15-23.
37. Worodria W, Massinga-Loembe M, Mazakpwe D, Luzinda K, Menten J, Van Leth

F, et al. Incidence and Predictors of Mortality and the Effect of Tuberculosis
Immune Reconstitution Inflammatory Syndrome in a Cohort of TB/HIV Patients
Commencing Antiretroviral Therapy. J Acquir Immune Defic Syndr2011;
58(1):32-7.
38. MacPherson P, Dimairo M, Bandason T, Zezai A, Munyati SS, Butterworth AE, et

al. Risk factors for mortality in smear-negative tuberculosis suspects: a cohort
study in Harare, Zimbabwe. Int J Tuberc Lung DisOctober, 2011; 15(10):1390-6.
39. Zachariah R, Fitzgerald M, Massaquoi M, Acabu A, Chilomo D, Salaniponi FML,

et al. Does antiretroviral treatment reduce case fatality among HIV-positive
patients with tuberculosis in Malawi? INT J TUBERC LUNG DIS2007;
11(8):848-53.
40. Harries AD, Zachariah R, Lawn SD. Providing HIV care for co-infected
tuberculosis patients: a perspective from sub-Saharan Africa. International Journal
of Tuberculosis and Lung Disease2009; 13:6-16.
41. Franke MF, Robins JM, Mugabo J, Kaigamba F, Cain LE, Fleming JG, et al.
Effectiveness of Early Antiretroviral Therapy Initiation to Improve Survival
among HIV-Infected Adults with Tuberculosis: A Retrospective Cohort Study.
PLoS Med2011; 8(5):e1001029.
43
42. WHO policy on collaborative TB/HIV activities Guidelines for national
programmes and other stakeholders. Geneva, Switherland: World Health
Organization; 2012 (WHO/HTM/TB/2012.1).
43. Improving the diagnosis and treatment of smear-negative pulmonary and extra
pulmonary tuberculosis among adults and adolescents: recommendations for HIV-
prevalent and resource constrained settings. Geneva: World Health Organization;
2007.
44
11. ANEXES
ANEX I Data collection format
INTRODUCTION
This patient information collection format is intended to assess risk factors of mortality
in TB-HIV co-infected patients with and without ART in Bahirdar town governmental
health institutions. The study will be conducted through reviewing patient records in
TB and ART clinic and visiting the home/calling if the status of the patient is not
recorded or found in the ART follow up form and TB registration book. The study is
aimed to fill the information gap and provide evidence for planners, decision makers
and program implementers related to ART and TB activities.
Date______/______ /_______ Questionnaire code___________
Name of the health institutions______________
Name of data collector________________ signature_____
Name of supervisor____________________Date__________ signature__________
Part-I SOCIO DEMOGRAPHIC CHARACTERISTICS (to be filled from ART

form)
No Variables Coding categories

101 Year ______________
102 Address 1. Urban
2. Rural
9. Unknown
103 Age(years) ______________
104 Sex 1. Male
2. Female
9. Not recorded
105 Religion 1. Orthodox.
45
2. Muslim
3. Protestant.
4. Catholic
5. Others specify___________
9. Not recorded
106 Marital status 1. Single
2. Married
3. Divorced
4. Widowed
5. Separated
9. Not recorded
107 Educational status 1. Not educated
2. Primary
3. Secondary
4. Tertiary
9. Not recorded
108 Occupational status 1. Farmer
2. Merchant
3. Government employee
4. Non government employee
5. Day laborer
6. Driver
7. Commercial sex worker
8. Jobless
9. Other specify-----------
99. Not recorded
46
Part II: Base line Clinical, laboratory and ART information

201 Past opportunistic illness(list all 1. Zoster
mentioned)
2. Bacterial pneumonia
3. Thrush(oral/vaginal)
4. Diarrhea (chronic/acute)
5. Pneumocystis pneumonia
6. Ulcers (mouth/genital)
7. Extra pulmonary TB
8. Cryptococcus meningitis
9. Other (specify)……….
99. Not recorded

202 Past TB treatment 1. Yes If 2
and 9
2. No skip to
204
9. Not recorded
203 Past TB treatment type 1.2SRHZ/6EH
2.2HRZES/1HRZE/5HRE
3.2HRZE/6HE
4. 2SRHZ/4RH
5. Not determined
9. Not recorded
204 Past prophylactic medications 1. Not given
(check all)
2. Cotrimoxazole
3. Fluconazole
4. Other medication--------
9. Not recorded
205 Weight at base line __________Kg
206 Height at base line __________cm
47
207 Body mass index _________kg/m2
208 Functional status 1. Working
2. Ambulatory
3. Bed ridden
9. Not recorded
209 WHO clinical staging 1. Stage III
2. Stage IV
9. Not recorded
210 CD4 count ________ cells/µl
212 Hemoglobin level _______ mmHg
212 ALT _______
213 AST ______
214 Acid Fast Bacilli result during TB 1. Positive
diagnosis
2. Negative
3. Not determined
Part III: TB location and category

301 TB location site 1. Smear +ve PTB
2. Smear -ve PTB
3. Extra PTB(Specify____________)
9. Not recorded
302 TB type 1. New 4. Failure
2. Defaulter 5. Chronic
3. Relapse 6. Retreatment
7. Other 9. Not recorded

303 Treatment Regimen 1. Category I
2. Category II
3. Category III
9. Not recorded
48
Part-IV Anti-TB, ART and Prophylactic medication

401 Date of TB diagnosis ______/_____/_______
402 Date of anti-TB initiation ______/_____/_______
403 Confirmed HIV+(date) ______/_____/_______
404 Anti-TB regimen 1. 2HRZE/6HE
2. 2HRZES/1HRZE/5HRE
3. 2RHZ/6EH
4. 2HRZE/4RH
9. Not recorded
405 Missed days during DOTS ________________
406 ART treatment 1. Started If 2
skip
2. Not started to
412
407 Time of ART initiation 1. Before TB treatment
2. Less than 2 weeks
3. 2 weeks-2month
4. 2-4 month
5. Greater than 4 month

408 ART eligibility criteria 1. CD4<200
2. WHO stage IV
3. WHO stage III with TLC<1200
4. Not recorded
409 Date of ART initiation ___/_____/_______
410 ART Regimens given 1. la=d4t-3TC-NVP
2. lb=d4t-3TC-EFV
3. lc=AZT-3TC-NVP
4. ld=AZT-3TC-EFV
5. Tdf-3TC-NVP
49
6. Tdf-3TC-EFV
7. 2nd line regimens

411 Adherence to ART drugs 1.Good
2.Fair
3.Poor
9.Not recorded
412 CPT prophylaxis 1. Prescribed before TB treatment If 1
and 3
2. Prescribed during TB treatment
Skip
3. Not started to 413
9. Not recorded
413 Date of initiation of CPT _______/_______/_______
414 Flucaonazole treatment 1. Prescribed before TB treatment
2. Prescribed during TB treatment
3. Not recorded
Part V Patient’s follow up information (to be filled from ART follow up form and
TB logbook).Please document the current or the recent results.

501 Latest follow up date _____/____/_______ For
502 Duration since initiation of ART ___________(months) ART
started
503 Recent weight in KG ______________
504 Recent functional status 1. Working
2. Ambulatory
3. Bedridden
4. Not recorded
505 Recent WHO staging 1. Stage III 2. Stage IV
506 Recent Opportunistic infections 1. Zoster
2. Bacterial pneumonia
50
3. Thrush(oral/vaginal)
4. Diarrhea (chronic/acute)
5. Pneumocystis pneumonia
6. Ulcers (mouth/genital)
7. Extra pulmonary TB
8. Cryptococcus meningitis
9.Other
(specify)……………………
507 ART Drug side effect 1. No 2. Nausea
3. Diarrhea 4. Fatigue
5. Headache 6. Numbness
7.Rash 8.Anemia
9. Fat change 10. Night mare
11. Dizziness 12. Others

508 Recent CD4 count _______________
509 Outcome of TB treatment 1. Cure
2. Treatment completed
3. Default
4. Failure
5. Death
510 Date of time of completion _________/_________/_______
Part VI: Base line Social condition
NO Variable Coding category

601 Religious/supportive 1. Yes
2. No
9. Not recorded
51
602 HIV serostatus 1. Wife/husband 5. Relatives
disclosure 2. Own child 6. Nobody knew
3. Parent(s) 7. Others specify___
4. Brother(s)/sister(s) 9. Not recorded

603 Condition of the 1. Health Ask if
husband/wife Married
2. Chronically ill and skip
to 608 if
3. Dead 3
4. Unknown
9. Not recorded
604 HIVstatus of husband 1. Not asked If 1,3,4
or wife skip to
2. Positive 606
3. Negative
4. Unknown
9. Not recorded
605 Was/is on ART 1. Yes
2. No
9. Not recorded
606 TB status of husband 1. Not asked If 1,3,4
or wife skip to
2. Positive 606
3. Negative
4. Unknown
9. Not recorded
607 Was/is on TB Rx 1. Yes
2. No
9. Not recorded
608 General concern 1. Financial issue 6. HIV status disclosure
identified
2. About the children 7. Adherence to
52
treatment
3. Marital relationship 8. Dietary problems
4. Family relations 9. Other specify__
5. Bereavement/grief 99. Not recorded
53
Annex II. Conceptual framework
Socio-demographic risk Base line clinical and

factors laboratory risk factors
 Age, Sex Past OIs
 Educational status WHO staging
 Religion, CD4 count,

employment Hemoglobin
 Marital status Total lymphocyte count
TB location site and

Fast progression
category
of the disease and
Smear +ve pulmonary complication -Medication and
TB regimen type
Smear _ve pulmonary -Anti-Tb treatment
TB
- ARV treatment
Extra pulmonary TB
- Cotrimoxazole
New, relapse, defaulter,
treatment
failure, chronic
-ART&Anti-TB
regimen
Death
54
DECLARATION
I, the undersigned, declare that this thesis is my original work and has not been
presented for a degree in this or another university and all the sources of materials used
for the thesis have been fully acknowledged.
Name: Balewgizie Sileshi
Signature: ______________
Date __________________
This thesis work has been submitted for the examination with my approval as a
university advisor
Name: Dr. Nigussie Deyessa (MD, MPH, PhD)
Signature: ____________________
Date: ________________________
55

Balewgizie Sileshi

Uploaded by

Copyright:

Available Formats

Balewgizie Sileshi

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Balewgizie Sileshi

Uploaded by

Copyright:

Available Formats

ADDIS ABABA UNIVERSITY

COLLEGE OF HEALTH SCIENCES

SCHOOL OF PUBLIC HEALTH

PREDICTORS OF MORTALITY AMONG TB-HIV CO-INFECTED

PATIENTS DURING TB TREATMENT IN BAHIRDAR TOWN

GOVERNMENTAL HEALTH INSTITUTION, ETHIOPIA.

A RETROSPECTIVE COHORT STUDY