Table 1. Results for studies on preventive therapy in HIV-infected individuals and in household contacts.

Appraisal	HIV Infection, n = 59		Household Contacts, n = 14		Total, n = 73
	Major Category	Minor Category	Major Category	Minor Category	Major Category	Minor Category
Year of publication
1990–1995	2		0		2
1996–2000	11		1		12
2001–2005	11		2		13
2006–2012	35		11		46
Objective
Effects on health outcomes	44		3		47
Evaluation of IPT for preventing TB or progression to AIDS	33		3		36
Comparing IPT versus no IPT only		11		0		11
Comparing various regimens and dosing schedules		9		0		9
Comparing various durations		2		0		2
Comparing various patient groupsa		1		0		1
Comparing IPT versus HAART with or without IPT		4		0		4
No comparison		6		3		9
Frequency of and risk factors for adverse effectsb	31		3		34
Drug resistance among TB cases during or after IPTb	6		0		6
Delivery	35		13		48
Evaluation of IPT completion/adherence ratec
Comparing different regimens		2		1		3
Comparing interventions for enhancing completion and/or adherence		1		0		1
Frequency of and risk factors for non-completion or non-adherenced		28		6		34
Comparison of various IPT enrolment methods		1		1		2
Barriers to implementation		1		0		1
Assessment of practices		2		5		7
Cost-effectiveness	4		0		4
Study design
Comparative studies	31		2		33
Individually randomized trials		19		0		19
Group-randomized trials		0		0		0
Non-randomized cohort comparisons		10		1		11
Before–after comparisons		1		0		1
Other		1		1		2
Non-comparative studies	28		12		40
Prospective cohort studies		19		4		23
Retrospective cohort studies		5		5		10
Cross-sectional clinical studies		2		1		3
Surveys		2		2		4
Setting
Estimated TB incidence per 100,000 population (2005)	57e		14		71e
<50		2		1		3
50–99		5		2		7
100–299		7		3		10
≥300		43		8		51
Estimated HIV prevalence among TB patients (2005)	57e		14		71e
<5%		3		4		7
≥5%		54		10		64
Study location	59		14		73
Research setting		22		0		22
Mixed research – routine setting		10		2		12
Routine setting		27		12		39
Generalizability
Study results are generalizable:	59		14		73
Irrespective of epidemiological or health care setting		31		4		35
To similar epidemiological or health care settings		18		7		25
Not beyond national/local setting		2		2		4
Other
Small sample size		3		0		3
Assessment of operational issues in research setting		5		1		6
Effectiveness versus efficacy
Methods aimed at establishing (relevant studies)	44f		3f		47f
Efficacy		16		0		16
Effectiveness		12		3		15
Mixed		16		0		16

^aFor example, comparing patients with positive versus negative tuberculin skin tests.

^bIncluding studies that addressed effects on treatment outcomes.

^cStudies testing a hypothesis about measures to improve treatment completion or adherence.

^dAs specific study objective, no hypothesis testing about measures to improve treatment completion or adherence.

^eTwo multi-country studies situated at locations with different TB incidences and HIV prevalence; one of these in various regions.

^fNumber of studies evaluating effects on health outcomes.

HAART, highly active antiretroviral treatment.