Preventive treatment for tuberculosis has been linked to a 61% reduction in the risk of developing the disease and a 45% reduction in mortality in people living with HIV.

Preventive treatment for tuberculosis (TB) is effective not only in carefully controlled clinical trials but also in routine healthcare systems in high-burden countries. This is the conclusion reached by the authors of a large international study, PROTECT, published in The Lancet HIV. The analysis included 235,424 people living with the human immunodeficiency virus (HIV) from six countries: Haiti, Kenya, Nigeria, Uganda, Ukraine, and Zimbabwe.

The main result was quite significant. Initiating preventive treatment for tuberculosis soon after enrollment in the human immunodeficiency virus (HIV) care program was associated with a 61% relative reduction in the risk of developing tuberculosis. The pooled odds ratio was 0.39, with a 95% confidence interval of 0.34 to 0.43. At the same time, the risk of death was approximately 45% lower: the pooled odds ratio was 0.55, with a 95% confidence interval of 0.51 to 0.58.

The study is particularly important because the authors studied unselected volunteers receiving treatment under ideal research conditions. The source of their data was data from routine care programs for people with human immunodeficiency virus (HIV). Thus, the researchers attempted to answer a practical question: does the preventive effect persist in real-world settings, where patients vary by age, stage of infection, immune status, country of residence, and access to medical care.

The findings support the strategy of integrating tuberculosis prevention directly into HIV treatment programs and initiating it early, if possible. The authors emphasize that the effect was observed in significantly different epidemiological settings and among different patient groups, making the results particularly relevant for global programs combating both infections simultaneously.

Why is tuberculosis especially dangerous for people with HIV?

Tuberculosis develops after infection with the bacterium Mycobacterium tuberculosis, but infection alone does not necessarily mean active disease. According to the World Health Organization, approximately a quarter of the world's population has been infected with the tuberculosis pathogen at some point. In most people, the immune system keeps the bacteria under control, and clinical disease does not develop. In the general population, the lifetime risk of developing an active infection is estimated at approximately 5-10%, but in immunocompromised individuals, the situation changes dramatically.

The human immunodeficiency virus (HIV) targets precisely those cellular immune mechanisms that are critical for controlling mycobacteria. Therefore, an infection that could have existed in the body for years without symptoms now has the opportunity to reactivate. The US Centers for Disease Control and Prevention (CDC) estimates the risk of developing active tuberculosis in people with the human immunodeficiency virus (HIV) to be approximately 12 times higher than in people without the infection.

An additional problem is that tuberculosis and the human immunodeficiency virus (HIV) mutually reinforce each other's negative effects. In severe immunodeficiency, tuberculosis more often takes extrapulmonary and disseminated forms, can present atypically, and is more difficult to diagnose. This is why tuberculosis has remained one of the most significant causes of death in people with HIV for decades, despite the tremendous progress of antiretroviral therapy.

From a prevention perspective, this creates an unusually favorable opportunity. If active tuberculosis has not yet developed, but there is a risk that the mycobacteria present in the body will subsequently cause disease, anti-tuberculosis drugs can be administered early. The combined use of antiretroviral therapy and preventive treatment for tuberculosis addresses two aspects of the problem simultaneously: the former restores immune defenses, and the latter reduces the likelihood that remaining mycobacteria will cause active disease.

Why prevention is especially important for HIV

Factor	What's happening	Practical significance
Weakening of cellular immunity	It is more difficult for the body to control Mycobacterium tuberculosis	The likelihood of the infection developing into active tuberculosis increases
High risk of tuberculosis	People with HIV have a significantly higher risk	Prevention offers especially great potential benefits
Concomitant use of antiretroviral therapy	The immune system is gradually recovering.	Complements the action of anti-tuberculosis prophylaxis
The ability to prevent disease before symptoms appear	Treatment is carried out before the development of active tuberculosis.	Serious illness and some deaths can be prevented
High burden of HIV and tuberculosis co-morbidity	Particularly pronounced in a number of countries in Africa and other regions	Mass, not just individual, preventive programs are needed

Source: World Health Organization and U.S. Centers for Disease Control and Prevention. [1]

More than 235,000 patients from six countries: how the PROTECT study was conducted

The study included 235,424 people with human immunodeficiency virus (HIV). The largest group was from Haiti (69,936 people), followed by Uganda (65,936), Ukraine (57,851), Zimbabwe (24,099), Kenya (9,549), and Nigeria (8,053). Thus, the authors were able to compare the effectiveness of a single prevention strategy in regions with very different health systems and TB epidemiology.

Most of the participants were of economically active age: 147,667 people, or 62.7%, were aged 25 to 44 years. The database included 133,582 women, accounting for 56.7% of the sample, and 101,842 men, accounting for 43.3%. The large cohort size allowed the researchers to evaluate the effectiveness of prevention not only on average but also to test whether the effect is consistent across different epidemiological conditions and patient groups.

PROTECT, however, was not a classic randomized trial, in which the investigator randomly assigns one patient a drug and another a control. The authors used routinely collected individual electronic medical data and employed a method known as targeted clinical trial emulation. Using predefined rules, the researchers attempted to replicate in an observational database the question that might be posed in a randomized trial: what happens if preventive treatment is initiated shortly after enrollment in a human immunodeficiency virus (HIV) care program, compared to not initiating treatment.

The main analysis used an eight-week period during which patients could initiate prophylaxis therapy after entering the care system. This approach is important because, in real life, medication is almost never prescribed to all patients on exactly the same day. The authors also conducted analyses with different prophylaxis initiation windows, from earlier to longer ones, to test the robustness of the results to the chosen definition of treatment initiation.

Who participated in PROTECT?

Country	Number of participants
Haiti	69,936
Uganda	65,936
Ukraine	57,851
Zimbabwe	24,099
Kenya	9,549
Nigeria	8,053
Total	235,424

Source: PROTECT publication in The Lancet HIV. [2]

The risk of tuberculosis was reduced by 61%

The main outcome of the study concerns the development of tuberculosis after the start of follow-up. Among patients who began preventive treatment within the study period, the pooled odds ratio for developing tuberculosis was 0.39 compared to the group without such treatment. The confidence interval (CI)—from 0.34 to 0.43—was significantly less than one, indicating a statistically significant association between preventive treatment and a reduced risk of developing the disease.

Translating this statistical indicator into more understandable terms, we can say that the relative risk of developing tuberculosis was approximately 61% lower. However, it is important to avoid a common misconception here: this does not mean that prophylaxis prevents tuberculosis in 61 out of every 100 patients. What is at issue is the relative difference between the compared groups over time; the absolute benefit for a given individual depends on their baseline risk of developing the disease.

This distinction is crucial. If the baseline probability of disease in a given group is very high, a relative reduction of 61% can prevent a large number of cases. If the baseline risk is significantly lower, the absolute number of cases prevented will be lower, although the relative effectiveness may remain similar. Therefore, the PROTECT results primarily demonstrate the effectiveness of a preventive strategy at the population level of people with HIV, rather than allowing a single figure to predict the individual prognosis of any given patient.

It is particularly important that the protective association was observed across multiple countries with varying TB prevalence and different characteristics of their healthcare delivery systems. This distinguishes PROTECT from a small, single-center study: if a similar effect is found in Haiti, Ukraine, Uganda, Zimbabwe, Kenya, and Nigeria, the likelihood that the strategy is applicable only to a single, narrow population becomes significantly less likely. However, the magnitude of the effect could have varied across countries, so the pooled effect should be considered a summary estimate.

Key results of PROTECT

Exodus	Combined risk ratio	95% confidence interval	Simplified interpretation
Development of tuberculosis	0.39	0.34-0.43	Approximately 61% lower relative risk
Mortality	0.55	0.51-0.58	Approximately 45% lower relative risk

These percentages reflect relative, not absolute, risk reduction.[3]

Preventive treatment was also associated with a 45% reduction in mortality.

Even more important from a clinical perspective was the mortality analysis. The pooled odds ratio for death was 0.55, meaning those receiving prophylactic treatment had a roughly 45% lower risk of death during follow-up. The 95% confidence interval ranged from 0.51 to 0.58, indicating a sufficiently accurate statistical estimate for the entire large cohort.

Moreover, significant heterogeneity was observed between countries: for the mortality analysis, the statistical heterogeneity index I² reached 96%. This is an important detail because it demonstrates that the effect size was far from uniform across all six national programs. Differences may be related to the baseline condition of patients, the prevalence of tuberculosis, the timeliness of diagnosis, the availability of antiretroviral therapy, and other characteristics of the healthcare systems.

The reduction in overall mortality could potentially be explained by several mechanisms. The most obvious is the prevention of tuberculosis itself, which can quickly become a severe or fatal disease in people with severe immunodeficiency. Furthermore, timely preventive treatment is usually part of a more organized patient management program: the patient regularly interacts with the healthcare system, receives antiretroviral therapy, and is assessed for tuberculosis. However, the observational design of the study does not allow us to conclude that the entire 45% difference in mortality is due solely to the direct pharmacological effect of the preventive drug.

This is why the statement "preventive treatment was associated with a reduction in mortality" is scientifically more accurate than the statement that the drug itself reduced mortality by exactly 45%. Nevertheless, the combination of a large sample size, a consistent reduction in tuberculosis risk, and the observed association with mortality provides compelling programmatic arguments in favor of prophylaxis among people with human immunodeficiency virus. The authors of the final publication also conclude that preventive therapy effectively reduced tuberculosis incidence and mortality across various epidemiological settings and risk groups.

Why it's important to start preventative treatment early

One of the most practically significant findings from PROTECT relates not only to the question of "to treat or not to treat" but also to the timing of prophylaxis initiation. The main analysis examined the initiation of therapy within the first eight weeks of enrollment in the HIV care program. According to the research team, the most attractive strategy is even earlier initiation, in the absence of contraindications and after active tuberculosis has been ruled out.

The logic here is simple: the period following the detection of the human immunodeficiency virus or the initiation of specialized medical care may coincide with a particularly high risk of undetected or rapidly developing tuberculosis. The longer prophylaxis is delayed, the longer the patient remains without additional protection. In the materials accompanying the PROTECT publication, the researchers note that initiation within the first two weeks is particularly desirable, although the significant effect was also maintained when using an eight-week window.

However, the "earlier, the better" principle doesn't mean that an anti-TB drug should be automatically prescribed at the first contact with a doctor. Before prophylactic treatment, it's essential to ensure that a person does not have active TB. If the disease already exists, a single prophylactic drug or an incomplete regimen will not be sufficient: the patient requires a full combination of drugs to treat the active infection. This is one reason why prevention programs are closely linked to high-quality TB screening.

From a programmatic perspective, these results suggest that TB risk assessment should not be postponed until a separate visit several months later. It is far more effective to integrate it into the same pathway through which patients receive HIV diagnosis confirmation, testing, antiretroviral therapy, and follow-up care. This makes prevention part of the standard medical care package, rather than an additional procedure that some patients may simply not access.

What an integrated approach might look like

Help stage	What needs to be done	For what
Inclusion in the HIV treatment program	Assess the symptoms and risk of tuberculosis	Don't miss an existing active disease
Exclusion of active tuberculosis	Conduct the necessary diagnostics according to clinical indications	The prophylactic regimen is not intended for the treatment of active disease.
Initiation of antiretroviral therapy	Control the human immunodeficiency virus	Restoring immunity further reduces the risk of tuberculosis
Early initiation of prophylaxis	If possible, do not delay after establishing the indications	The period during which the patient remains unprotected is reduced
Observation	Monitor medication intake and symptoms	Allows to assess tolerance and detect the disease in a timely manner

The overall logic is consistent with the approach evaluated in PROTECT and with recommendations for combining care for HIV and TB. [4]

How does prophylactic treatment work and why isoniazid was frequently used in the study?

In the national programs studied in PROTECT, a significant proportion of patients received a traditional isoniazid-based prophylaxis regimen lasting approximately six months. Isoniazid is a key anti-tuberculosis drug that can suppress mycobacteria before infection leads to clinically evident tuberculosis. Therefore, people take the drug not because they necessarily already have active tuberculosis, but to prevent its development.

This is fundamentally different from the treatment of active tuberculosis. Once the disease has been established, a combination of several anti-tuberculosis drugs is required, as using an insufficient regimen does not provide complete treatment and creates additional complications. Preventive therapy is intended for people whose active tuberculosis has been ruled out, but whose risk of developing it is still high—and human immunodeficiency virus (HIV) is one of the most significant risk factors.

In recent years, shorter prophylactic treatment regimens have emerged, so the PROTECT results should not be interpreted as proof that six months of isoniazid is necessarily the only or optimal option for every patient. Rather, the study demonstrates a more fundamental principle: the mere implementation of a prophylactic treatment program in people with human immunodeficiency virus (HIV) produces significant results in real-world settings. The specific regimen is determined by national guidelines, drug availability, drug interactions, and patient characteristics.

When prescribing isoniazid, the physician also considers potential adverse reactions, primarily effects on the liver and peripheral nerves, as well as the concomitant use of other medications. This further demonstrates why prevention cannot be reduced to mass self-administration of antibiotics: it requires a clinical assessment, exclusion of active tuberculosis, and subsequent medical monitoring. PROTECT evaluates the effectiveness of organized medical programs, not self-administered preventive medications.

How PROTECT differs from a typical clinical trial

One of the study's strengths is also its main methodological limitation. This is a prospective cohort study with meta-analytic data pooling, based on real-world programmatic practice, rather than a randomized controlled trial. Therefore, it allows us to see what happens to hundreds of thousands of ordinary patients, but it cannot fully guarantee that the groups receiving and not receiving prophylactic treatment were initially comparable.

For example, a patient who was promptly prescribed prophylaxis could potentially differ from someone who was not. They might attend clinic visits more frequently, adhere better to antiretroviral therapy, have different immune indicators, or receive care at a more organized medical center. Researchers use statistical methods to account for such differences as much as possible, but observational data always contain the possibility of residual confounding. This limitation should be considered when interpreting effect sizes.

The targeted trial emulation method is designed to reduce some of the typical biases of observational studies. The authors formulate inclusion criteria in advance, define the treatment strategies to be compared, the observation start time, and the treatment assignment time window, as if designing a hypothetical randomized trial. These rules are then applied to existing data. This approach significantly increases analytical rigor, but statistical simulation still does not transform observed medical practice into a truly random treatment assignment.

On the other hand, this design answers a question often under-reported in clinical trials: what happens after scaling up the strategy nationally. Patients in PROTECT were treated in very different countries, and healthcare workers faced common challenges—missed appointments, differences in laboratory testing, uneven resources, and variability in adherence to recommendations. Therefore, discovering a significant preventive effect in these specific settings is of great value to healthcare planners.

Strengths and limitations of the study

Peculiarity	Why is this important?
235,424 participants	Very high statistical power
Six countries	The ability to test the strategy in different epidemiological conditions
Real medical programs	Results closer to everyday practice
Individual electronic health data	Allows analysis of treatment timeframes and outcomes
Target test emulation	Reduces a number of systematic errors in observational studies
Lack of true randomization	Unmeasured differences between groups cannot be completely ruled out
Marked heterogeneity in mortality	The effect size varied across countries
Differences in national health care systems	At the same time, they increase the applicability of the results and complicate the comparison

Source: PROTECT design and results.

Why the findings matter for the global fight against tuberculosis

The World Health Organization continues to rank tuberculosis among the world's leading infectious causes of death. Globally, approximately 10.7 million people were diagnosed with tuberculosis in 2024, resulting in approximately 1.2 million deaths. People with human immunodeficiency virus (HIV) are among the groups at highest risk of severe disease and death, so preventing the disease in this population has the potential to significantly impact global statistics beyond the number of patients treated.

Scaling up prevention has already occurred in recent years. Following the strengthening of preventive treatment programs, including with support from the US President's Emergency Plan for AIDS Relief, millions of people living with the human immunodeficiency virus (HIV) have received preventive anti-TB drugs. The PROTECT research team notes that since 2018, more than 13 million people living with the human immunodeficiency virus (HIV) have received preventive treatment. This new study provides evidence that such large-scale investments are yielding measurable results in real-world practice.

This also has significant implications for future resource allocation decisions. Clinical trials have already demonstrated the effectiveness of preventive regimens, but governments and international organizations need to know whether the benefits persist after transferring the strategy to public clinics, where conditions are much less controlled. PROTECT shows that prevention remains associated with significant reductions in both TB incidence and mortality, even when analyzing large national programs.

The study's primary practical implication, therefore, is not the development of a new drug, but rather the validation of an existing medical intervention. If a person with human immunodeficiency virus (HIV) is promptly enrolled in care, active TB is ruled out, an appropriate prophylactic regimen is prescribed, and antiretroviral therapy is provided, a significant proportion of future TB cases and likely a significant proportion of deaths can be prevented. This is a rare example of an intervention in global health where the technology already exists, and the primary challenge is ensuring its timely implementation.

What do the results mean for the patient?

The PROTECT results do not mean that every person with HIV should begin taking isoniazid or another anti-TB drug on their own. First and foremost, a medical examination for active TB is necessary. The presence of cough, fever, night sweats, unexplained weight loss, or other suspicious symptoms may require further diagnostic testing, as a full treatment regimen, rather than a prophylactic regimen, is used for advanced TB.

Preventive therapy should not be considered a substitute for antiretroviral treatment. On the contrary, maximum protection is achieved through a combination of strategies. Antiretroviral drugs suppress the replication of the human immunodeficiency virus and allow the immune system to recover, while anti-tuberculosis prophylaxis directly reduces the likelihood of developing the disease caused by mycobacteria. Combining these measures is the basis of modern tuberculosis prevention in people with human immunodeficiency virus.

Furthermore, a 61% relative risk reduction does not mean complete protection. Tuberculosis can still occur after preventive therapy, especially if a person continues to be exposed to the infection in a high-incidence region or has severe immunodeficiency. Therefore, the appearance of symptoms after completing prophylaxis still requires a visit to a doctor and examination.

Nevertheless, the magnitude of the observed effect makes the study's conclusion quite clear. For people living with the human immunodeficiency virus, preventive treatment for tuberculosis is not a secondary, additional measure, but an essential component of medical care. PROTECT demonstrates that the timely implementation of this strategy in routine programs across various countries is associated not only with a reduction in tuberculosis diagnoses but also with a significant reduction in mortality.

Key figures from the study

Indicator	Result
Total number of participants	235,424
Number of countries	6
Countries	Haiti, Kenya, Nigeria, Uganda, Ukraine, Zimbabwe
Age 25-44 years	147,667 (62.7%)
Women	133,582 (56.7%)
Men	101,842 (43.3%)
Main window for starting prevention	up to 8 weeks
Odds ratio for tuberculosis	0.39 (0.34-0.43)
Relative reduction in the risk of tuberculosis	≈61%
Mortality hazard ratio	0.55 (0.51-0.58)
Relative reduction in mortality risk	≈45%
Type of study	Prospective cohort, targeted trial emulation, and pooled analysis

Source: PROTECT publication in The Lancet HIV.

News source

Primary scientific publication:

Shah NS, Mishara F, Zissette S, et al.; PROTECT Study team. Programmatic effectiveness of tuberculosis preventive treatment for people with HIV in six high-tuberculosis-burden countries (PROTECT): a prospective cohort study and meta-analysis. The Lancet HIV. 2026.

DOI: 10.1016/S2352-3018(26)00140-2.

The authors' key finding: preventive treatment for tuberculosis in real-world programs for people living with human immunodeficiency virus (HIV) was associated with a significant reduction in tuberculosis incidence and mortality in six countries with varying epidemiological situations. This makes PROTECT one of the largest-scale studies demonstrating the benefits of tuberculosis prevention not only in clinical trials but also after its implementation in routine healthcare.