Back to contents

HIV and TB in Practice for nurses: TB symptom screening and linking to diagnosis and care

Theo Smart
Published: 16 July 2012

This edition of HATIP is kindly supported by the STOP TB department of the World Health Organization.

The previous edition of HIV and TB in Practice for Nurses focused on case finding strategies that are used in the clinic or in outreach strategies to hunt for TB.  This double edition looks at the use of symptoms to screen patients for TB. How do healthcare workers recognise a possible case of TB in the outpatient waiting room? Do outreach teams look for TB the same way? How do we make certain that the person who may or may not have TB makes it into diagnosis and care instead of being lost in the referral process?

Symptom screening is the only practical way to begin active TB case finding in low resourced settings. But the symptom screen used depends on whether the main focus of the screening exercise is to prevent TB transmission, or to improve someone’s health (and prevent transmission). Furthermore, symptom screening for members of the general population is different from the WHO 4 symptom screen used in people living with HIV and others at elevated risk of TB (TB contacts, etc). And finally, TB screening is different in adults than in children (diagnosis in children will be covered in an upcoming issue). Guidelines also vary from country to country,

But two things are important to know first of all:

1) People living with HIV can have TB without ever getting a chronic cough (≥ 2 weeks). If the TB symptom checklist used in your facility to screen people living with HIV requires that they have a chronic cough before doing lab tests, and other investigations for TB, then the facility is missing a lot of TB in people living with HIV.

2) Nevertheless, chronic cough is still a very important symptom to watch for because it usually means that a TB case is more infectious — this makes it an important symptom to screen for when the primary goal is to prevent TB transmission.

The importance of chronic cough

The importance of screening for chronic cough, to find and treat the most infectious cases, and reduce transmission in the community was demonstrated by the DetecTB active case finding study in Zimbabwe (DETECTB, London School of Hygiene & Tropical Medicine). Researchers were evaluating two different TB screening approaches in a community near Harare.

In one strategy, where outreach teams went door to door asking whether anyone in the home had a chronic cough. If they did, the team would collect their sputum specimens and send them to the lab for smear microscopy.

The other strategy involved advertising that a mobile van would be coming to the community offering TB screening services to anyone who came to them with symptoms of TB.

Microscopy was used rather than new more sensitive tests such as Xpert MTB Rif for TB diagnosis because people with smear positive pulmonary TB are highly infectious, and the researchers wanted to see if detecting, rapidly diagnosing, and providing prompt treatment and a cure for more of these really infectious cases would reduce TB transmission in the community. It did, by around 40% at the end of the trial. Over several years, that ought to reduce the prevalence of TB.

The study also showed that finding an infectious case early can have a powerful effect that may take a while to see. When the study data were first being analysed, it was clear that the mobile van services strategy detected a lot more TB cases than the door-to-door strategy. But in the final analysis, it was the door to door arm that appeared to have had the greater effect upon the spread of TB. The door to door approach could have found a higher percentage of the most infectious cases in the area being screened. The researchers stressed the door to door active case finding approach found these TB cases much earlier. This made it possible to put them on treatment before their TB had the opportunity to spread to many other people.

Even with a mobile van offering free TB diagnosis in the neighbourhood, people may be too busy, or are scared of being stigmatised, and probably don’t understand how serious their illness is. While they put off getting the health services they need, they do not know that they are exposing people, including their loved ones, to TB. It is probably more difficult for people to delay when someone is standing at their door with a sputum cup.

Finding people with a cough, getting them diagnosed and treating TB earlier prevents the spread of TB

Similarly, the sooner healthcare workers find out that a person with TB or the symptoms of infectious smear-positive TB has arrived at the waiting room or health facility, the sooner they can take steps to reduce opportunities for exposure to TB, and transmission to other patients, or to healthcare workers themselves.

First come, first served is the norm in many clinics in sub-Saharan Africa, but it is not as safe a way to manage outpatients, from an infection control standpoint, as cough triage.

Triage means identifying people who have TB or who could have TB as quickly as possible when they come to a health facility. This can be done by training a staff member/s such as a counsellor, or receptionists, or community health care workers, to serve as ‘cough monitors’ and having at least one designated cough monitor on duty at all times while the facility is open.

Where the screening takes place will depend on the design of the facility. Optimally, TB or cough screening should occur before a person can enter into an enclosed part of the facility, in particular, the waiting room.

Cough monitors or infection control officers should be trained to ask whether the person has been coughing, and if so how long (most programmes focus on chronic cough as the relevant symptom for infection control). They should also ask whether the person is a TB patient on treatment (who may still be infectious, especially if treatment only recently began) or if they are undergoing diagnostic evaluations for TB (and perhaps even picking up TB test results).

Those identified with cough should be taught proper cough etiquette and respiratory hygiene and be given a surgical mask or cloth to cover their mouth with when they cough.

Good infection control practice dictates that people with a chronic cough be separated from other patients, if possible, and wait in a well-ventilated area. It may be best if the waiting rooms are outside or in a partly-ventilated, semi-enclosed area. If there is only one general waiting room, and it is in an enclosed area without good air circulation, then people with TB or under evaluation with TB should be seated in the part of the waiting room with the best possible ventilation away from the other clients and staff.

Fast-track the diagnostic evaluation of anyone with chronic cough and move them, along with the returning patients with smear-positive TB and those under evaluation, awaiting diagnosis for TB, to the front of the queue for any other services needed to minimise the time that they spend in the facility.

Care should be taken not to stigmatise coughing patients, which may be easier if the clinic has an outdoor waiting area for everyone. Some clinics, such as the Ubuntu clinic near Cape Town have chosen to treat every patient as someone with potentially infectious TB, teaching everyone cough etiquette and making everyone wear surgical masks to cover their mouth while they are at the facility.Triage can have other benefits besides improving infection control. For instance, one resource, CREATE’s Case-based Curriculum on the 3 I’s, recounts an assessment of TB infection control challenges at Chimwemwe Clinic in Zambia, which resulted in a number of changes at the site:

  • the construction of individual waiting areas for TB and ART clients, and

  • the implementation of triage for cough upon arrival for all patients at the HIV clinic.

“The staff found that the move not only improved TB infection control measures but also increased the overall efficiency of providing TB and HIV therapy services,” wrote Eldred and Teeter.

Broader TB symptom screening in the general population

Not every facility performs triage, and asking about chronic cough won’t find every case of TB. This is most often the case in PLHIV. It can also be true in HIV- negative people, who can also have smear negative or extrapulmonary TB (EPTB), especially where there is poverty and malnutrition. In high-burden, resource-constrained settings, the number of these cases could be significant, and active case finding could lead to earlier case detection and improved health in these individuals as well.

In South Africa, it is now the policy to conduct intensified case finding in EVERY person who visits a health facility, even if that person has come to the health facility for reasons that have nothing to do with how they are feeling that day — for instance, TB screening should be performed in women who are bringing their child in for immunization and for men coming in for a medical circumcision.

While the WHO 4 Symptom screen may be useful in groups with very high rates of undiagnosed TB and HIV coinfection (like prisoners or people who inject drugs), it is not specific enough for TB to use in the general population. It could identify very large numbers of people as needing further diagnostic evaluations yet relatively few would actually have TB. How useful it is for screening in other groups at elevated risk of TB is unclear; the answer would depend upon burden of TB and the laboratory capacity within the setting. The risk of TB is consistently high among close household contacts of a person with smear-positive pulmonary TB, however, and South Africa useds the 4 symptom screen in such cases.

At present, health facilities in South Africa are supposed to ask patients if they have symptoms matching South African TB guideline definitions of a TB suspect:

  • Persistent cough for more than 2 weeks

  • Fever for more than 2 weeks

  • Drenching night sweats

  • Loss of appetite

  • Unexplained weight loss (more than 1.5 kg in a month)

  • A general feeling of illness (malaise) and tiredness

  • Shortness of breathe, difficulty breathing or chest pain

TB symptom screening for case detection in PLHIV

However, as has already been noted, PLHIV with TB often don’t have chronic cough — and if that is the only symptom that is being looked for, a lot of possible TB cases could be overlooked among PLHIV.

There are other symptoms of TB but these are either very non-specific (they could be symptoms of almost anything) or not consistently found, including constitutional symptoms (feeling generally ill), fatigue, chills, night sweats, fever, sometimes coughing with blood, chest pain, shortness of breath, weight loss, swollen lymph nodes and loss of appetite. This means that not every person with HIV-related TB is coughing up blood or has swollen lymph nodes.

For years, researchers did studies that combined these symptoms in checklists to try to find an easier way to identify who might have TB — without much luck. The checklists were either too general or too specific, missing cases.

Then WHO’s STOP TB Department came up with a TB symptom screen, based on a systematic analysis of the available clinical data. It cannot tell you if someone has TB, but it does a pretty good job of identifying which PLHIV do not have TB — which makes it possible to provide isoniazid preventive therapy (IPT) without much fear of undertreating an active case of TB.

Note: this symptom screen asks whether the person has been coughing, not whether the person has chronic cough (a classic symptom of TB). While it is important to remember that the longer someone has had a cough, the more likely that it could be caused by TB, many people living with HIV and TB never have a chronic cough. Some symptom screens ask for ‘any’ current cough in the past 24 hours, or longer, but some national guidelines use a screen that asks for a cough of more than 24 hours.

WHO guidelines recommend the use of the WHO 4-symptom TB screen every time HIV-positive people come into contact with the health system, scheduled or otherwise.

The WHO 4-symptom TB screen asks:

  • Have you had a cough in the past 24 hours?

  • Have you had any fever,

  • Night sweats

  • Or weight loss?

Some countries have added additional symptoms to the symptom screen to make it slightly more sensitive for conditions reported in their population. For instance, Namibia has a 5-symptom Screen, after adding swollen lymph nodes to the list. In some settings, ICAP’s screening tool also includes swollen lymph nodes and ‘pain or difficulty breathing.

If the answer for all four symptoms is “no”, TB can be ruled out with confidence, and the patient given IPT.

A case or two of TB might be missed, so WHO recommends repeating the symptom screen every time a PLHIV interacts with health services (including home visits)— even in people taking ART, and especially in people taking IPT, to eliminate any chance of a breakthrough case of TB going undertreated.

South African Department of Health TB screening tool

The 4 TB symptom screen is the first critical step in intensified case finding. It is essential that TB programmes and facilities focus first on how to do symptom-based screening well.

If the answer for any symptom on the 4 symptom screen is yes, WHO recommends investigating for TB and other infections in accordance with national guidelines and principles of sound clinical practice to identify either active TB or an alternative diagnosis.

Local guidelines do vary. Some programmes or facilities call for referral for TB diagnosis, or sputum specimens to be collected from every PLHIV with a positive symptom screen and sent off to the lab. Others do this only for symptomatic people with a cough and call for a clinical assessment of the rest. Some national programmes call for a clinical assessment first before referring or sending out specimens for diagnosis. What is done in practice may also depend upon the burden of disease in the region and local lab capacity. Nurses and other healthcare workers should check local guidelines.

It also pays to have good clinical common sense, and consider a range of factors from the patient’s other signs, symptoms, and general clinical condition. Any number of illnesses can also cause these symptoms and need to be diagnosed. For instance, in many parts of Africa and Asia, the most common cause of fever is malaria. 

If a person in your care is reporting symptoms, especially a person living with HIV, he or she needs a good clinical assessment to be sure they are receiving proper diagnosis and care, whatever their ailment, as many national guidelines recommend. For example, according to the South African guidelines, individuals whoscreen positive for symptoms of TB should undergo an evaluation by a doctor or nurse trained in TB diagnosis (and, then, if necessary further investigations, including labwork and chest x-rays as needed).

This may simply involve taking a good history, with special attention to:

  • duration of the symptoms,
  • TB risk factors,
    • whether the person has previously received TB treatment,
  • whether there is a history of TB exposure at home (and note, whether there are young children in the household should there need to be contact tracing),
  • whether they have worked in a congregate setting (like the mines);
  • whether they have other risk factors for cough (like smoking), or other obvious explanations for any particular symptom.

A physical examination and consideration of other diagnoses should proceed as usual.

“Someone with simple common cold can present with one or more of the TB symptoms, but probably does not need to go through the complete TB diagnostic workup,” Dr Haileyesus Getahun of WHO’s STOP TB department told HATIP. “Similarly, in the deep sea of TB in Guguletu, one symptom may give you a good indication for TB, but not in the slums of Addis,” he added.

Sound clinical practice should eliminate most of the cases that are obviously due to something else, and make certain that other important causes of illness are addressed. As for the other cases where the positive symptom cannot be easily explained after the clinical assessment, the case requires further evaluation. With the nurse clinician or doctor’s signoff, the diagnostic process can get underway, including whatever lab tests or radiography services are necessary. If TB is not thought to be likely, the patient should be managed appropriately and followed up to ensure that their symptoms have resolved.

Remember, in people living with HIV:

  • the risk of TB is very high

  • TB may not have the classical symptoms, and

  • the purpose of ICF is to identify TB early

Launching the diagnostic process: sputum collection

Depending on where the case was found, healthcare staff will either have to refer the person with presumptive TB somewhere else for TB diagnostic services (such as a TB unit or facility, or a microscopy centre/peripheral lab), or they will have to collect sputum and other specimens and send them to the lab.

The healthcare worker in charge of the person with presumptive TB will need some essential supplies, including referral forms, letters or patient-held cards if cases are referred out. Otherwise, he or she will need laboratory request forms, the sterile sealable sputum collection cup or vial provided by the laboratory service, and a safe sputum collection area.

It is very important that sputum collection be performed in a designated and well-ventilated sputum collection area, preferably outdoors, and away from other patients. If placed outdoors, a somewhat sheltered spot with running water so the patient can wash their hands afterwards would be best.  Sputum should never be collected in a bathroom or other enclosed area used by patients or staff. The healthcare worker should not be inside the sputum collection booth when the patient is coughing, particularly if it is indoors.

Reproduced from: The Sputum Collection Procedure Visual Guide, created for use in clinics in South Africa with support from USAID through the TASCII TB Project managed by University Research Co., LLC. Content developed by URC-CHS and Kwikpoint.

Step-by-step guide to sputum collection

TB screening and sputum collection for TB microscopy may have to be performed in a variety of healthcare settings. For example it may need to be performed in bedridden patients The Hospice Association of South Africa has published a guide on providing palliative care for patients with TB which includes the following instructions on safe sputum collection — including, in point 10, sputum collection in a bedbound patient.

1. Label the sputum bottle first with all the patient’s information.

2. The sputum should be collected in a well ventilated, dedicated area that is only used for sputum collection. An outside area can also be used as long as it has privacy and does not allow others to watch. Do not use toilets for sputum collection.

3. Explain the steps slowly and in brief points.

4. The person should first rinse out their mouth with water to remove any pieces of food.

5. Demonstrate how to give a deep cough from the bottom of the chest:

• Beginning with deep breathing.

• Cough forcefully.

• Repeat this three times before spitting into the bottle.

6. Give the patient the sputum bottle, without the lid to spit it into the bottle.

7. The patient must fit the sputum bottle underneath the lower lip and spit into the sputum bottle.

8. Ask the patient to take care not to spill sputum on the outside of the sputum bottle.

9. Don’t stand in front of a patient while s/he coughs.

10. If the patient is in bed at home: Move the bed until the patient is in front of an open window or door. The patient must turn their head towards the open window or door when coughing and spitting into the sputum bottle.

11. Supervise the collection, but do not stand in front of the patient.

12. Be ready to replace the lid on the sputum bottle immediately after the patient coughed up sputum.

13. Make sure the patient coughed enough sputum into the sputum bottle (about a teaspoon full).

14. Close the sputum bottle by turning the lid three times to ensure that it is completely closed.

15.  Wash your hands after closing the sputum bottle.

16. Take the sputum bottles as soon as possible to the health facility in a cooler bag or box. Don’t expose the sputum bottles to direct sunlight.

Wits University’s Reproductive Health Research Unit has produced a number of resources on infection control and symptom screening, which contain the following useful advice for patients on sputum collection: “If you cannot cough up sputum, try breathing steam from a hot shower or pan of boiling water first, and then cough.”

Preparing sputum samples: what is a good sputum sample?

To optimise the chances of diagnosing TB from a sputum sample, South Africa’s National Health Laboratory Service makes the following recommendations:

  • Submit specimens to the laboratory as soon as possible

  • If transport to the laboratory is longer than 1 hour specimens should be refrigerated

  • Characteristics of a good sputum specimen: Mucoid or mucopurulent appearance, minimum amounts of saliva 

  • Optimal volume: 5mL-10mL
  • Minimum volume: 2 mL

If the patient cannot produce enough sputum, a common problem in people with HIV (especially in children), an excellent new resource, Managing TB in a New Era of Diagnostics from Aurum Institute contains materials on the safe collection of induced sputum, and other specimens.

At present, some programmes require two specimens for smear microscopy and potentially one for culture. Not all countries have culture of course, and when they do, it may only be requested in smear negative cases. Policies also vary as to when and where the specimens should be taken — in some, the patient is requested to take a sputum specimen container home with them, and to give a specimen as soon as they wake up in the morning. There is of course a danger that the patient could get lost to follow-up and not return (see below). Similarly, in some districts in South Africa, where Xpert MTB/RIF testing is available, only one spot sputum specimen needs to be sent in, though other specimens will be needed when the client returns for their results.

Keep track of any PLHIV with a positive TB symptom screen!

It is important to make sure ICF doesn’t just stop with the symptom screen, when there is a clinical assessment before specimens are collected and sent to the lab, or somewhere in the process of referral or once specimens are sent to the lab. There needs to be a system in place to keep track of people while they are being evaluated for TB. These exist in some health systems but not all, or in some parts of the health system, but not others.

Health care facilities sending out specimens to a lab should have a register to keep track of specimens sent to the lab. Information on each specimen sent out should be recorded, including when it was sent out, and the register should include a column for recording the date that the lab results were received back at the facility, and possibly the date they were given to the patient. The lab specimen register should be reviewed regularly to make certain specimens or results have not gone missing. The register can also be used to keep watch for changes in lab turnaround.

The International Union Against Tuberculosis and Lung Disease (the Union) notes in its recently published programmatic guide Implementing Collaborative TB-HIV Activities:

“In some countries, registers are kept of ‘patients with chronic cough’ or patients with suspected TB. One purpose of this register is to ensure that the results of the diagnostic tests have been received and acted upon. These registers should also have a column that indicates the conclusion of the investigation: was the person who was identified through ICF found to have TB or not? If TB was confirmed, when did TB treatment begin? Since screening for TB may take place in several sections of the health facility, multiple registers may be needed.”

The register may also provide a place to enter thorough contact information for the person with presumptive TB. Aurum Institute’s new TB Toolkit stresses that the healthcare worker should “obtain accurate contact and alternate contact details to ensure follow-up In the event of default. Confirm contact details at each visit!!!”

“Take 2 or even 3 cell phones numbers for each TB suspect. Write their details as if it is your own family member who will go untraced and die unless you fill the register properly!” said Dr Krista Dong, one of our advisors who works in KwaZulu Natal, South Africa.

Information to follow-up on ‘patients with suspected TB ‘ may also be entered in the the TB Case and Follow-up register. For example, Aurum Institute’s new TB Toolkit contains the following checklist to consider when filling out a sample TB case identification and follow-up register:

What are key issues in completing the TB case identification and follow-up register (suspect register)?

  • Use NHLS barcode sticker to document specimen number
  • If testing was done by Xpert MTB/RIF indicate this in the ‘Comments’ section

  • NB: Patients diagnosed with Xpert MTB/RIF also need to have a smear taken to monitor treatment response
  • Offer all TB suspects HIV testing. An extra column can be added to record HIV result

  • Remember to complete the monthly summary sheet at the back of the TB Suspect Register

When ICF happens beyond the clinic, similar follow-up mechanisms need to be put in place. For instance, trained counsellors at different HIV testing and counselling sites are now screening for TB as well — but with very different results. In one programme in the Western Cape of South Africa, people who screened positive on the symptom screen were referred for further diagnostic evaluations — but few ever completed the referral.

In a different programme, community healthcare workers (CHW) were linked to a mobile HIV testing and counselling services also providing. Whenever anyone would test positive on the TB symptom screen or test positive for HIV one of the CHWs would accompany them to the health facility. This is a much more effective way to link the patient to diagnosis and potentially to care but requires an investment in partnering with community health workers, expert patients or community based organisations.

The NGO, Partners in Health is known for piloting much of this sort of work with expert patients, first in Haiti, where the expert patients were called “Accompagnateurs.”

The accompagnateurs were usually former TB and/or current (if PLHIV) patients, trained to provide support to other patients in their homes and communities, and to accompany them if needed to the clinic, whenever there was a problem.

Partners in Health has a pilot test-training guide that can be downloaded in English, Kinyarwanda, Kreyol, Sesotho, and Chichewa. It provides a thorough basic training for home-based carers and support-givers on HIV and TB, with exercises where expert patients act out the symptoms of TB. It contains some useful educational materials such as:

"The three types of people who should always be tested for TB?

  • Someone who lives with or spends a lot of time with people with TB.

  • People with the symptoms of TB,

  • People living with HIV.

Tell participants that because TB can spread very easily among people who are living in close conditions, if one person in a home or family has TB, everyone in the home or family should go to the clinic to get tested for TB. Even if someone just thinks she has TB, everyone in her home or family should go to the clinic to get tested. As accompagnateurs, you should bring or send them to the hospital or clinic to get tested for TB."

But what impressed us about the guide is how the expert patient is trained to be alert and sensitive to the needs of the patient and their family affected by HIV and TB — acting as the eyes and ears of a health system that is truly ‘caring.’

Another model of community engagement in TB-HIV services, in Thyolo District Malawi is described in a paper by Zachariah et al. Community members were responsible for referral of people suspected of having TB, sputum collection and transport, and defaulter tracing (directly observed treatment by ‘guardians’).1

Getting the results to the patient

If, after all this trouble, the person with a presumptive case of TB never get their results back, it has all been a waste.

“It is important for the health worker to ensure that the results of sputum smear microscopy are received and that patients are informed whether they have TB or not. If TB is diagnosed, correct TB treatment should be commenced without delay. This will rapidly reduce the infectiousness of the patient,” according to the Union’s Implementing Collaborative TB-HIV Activities: A Programmatic Guide.”.

However, when a test comes back negative, and the patient still has symptoms, there is a good chance that they may still have TB. Smear microscopy is not very sensitive for TB, and many cases are missed particularly among people living with HIV. Even new tests like Xpert MTB/RIF, which can diagnose many cases of the smear negative TB cases, still miss a lot of cases. A negative result is often not the final word, in a person living with HIV. If the lab results come back negative and the patient still has symptoms, it is important to keep looking for their illness, to keep investigating!!! A future issue will look more closely at the harder to diagnose TB cases {including smear-negative TB, and EPTB).


Zachariah R et al. How can the community contribute in the fight against HIV/AIDS and tuberculosis? An example from a rural district in Malawi. Transactions of the Royal Society of Tropical Medicine and Hygiene, 100:167–175, 2006.

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.
Community Consensus Statement on Access to HIV Treatment and its Use for Prevention

Together, we can make it happen

We can end HIV soon if people have equal access to HIV drugs as treatment and as PrEP, and have free choice over whether to take them.

Launched today, the Community Consensus Statement is a basic set of principles aimed at making sure that happens.

The Community Consensus Statement is a joint initiative of AVAC, EATG, MSMGF, GNP+, HIV i-Base, the International HIV/AIDS Alliance, ITPC and NAM/aidsmap

This content was checked for accuracy at the time it was written. It may have been superseded by more recent developments. NAM recommends checking whether this is the most current information when making decisions that may affect your health.

NAM’s information is intended to support, rather than replace, consultation with a healthcare professional. Talk to your doctor or another member of your healthcare team for advice tailored to your situation.