17.4 Reporting


The key evaluation tool for all forms of TB is the periodic report. It must be presented in a standardized manner in two parts: case enrolment and treatment outcomes. The data presented in the report comes from the TB register. It is generally completed by quarter for drug-susceptible TB and by semester for DR-TB.

Evaluation of interim and final treatment outcomes is a fundamental stage in the evaluation. This evaluation is done through a cohort analysis. A "cohort" is a group of individuals presenting certain common characteristics and undergoing the same events. In respect to the evaluation of TB patients, a cohort is represented by patients all put under treatment within a given period of time (usually a quarter for drug-susceptible TB and a semester for DR-TB). At the end of treatment, a final outcome is assigned to each patient (Table 17.1).

Notes:
– The number of patients in each group should, in principle, be identical to those registered for the same interval in the case enrolment part of the corresponding periodic report. If it is different, an explanation should be given (e.g., patients “interrupting before treatment” can be excluded from the outcome analysis).
– The outcomes of patients "transferred in" should not be included in the outcomes of the facility to which they were transferred. Their outcome results should be recorded in the facility that initially enrolled the patient in TB treatment.

17.4.1 Case detection and enrolment report for TB

The elements necessary for defining a TB case (treatment history, bacteriological status, anatomical site of the disease, and HIV status) are defined in Chapter 7.

See Quarterly report for case enrolment, Appendix 32.

Main indicators

Proportion of confirmed pulmonary TB (PTB)
= Number of PTB cases confirmed enrolled/Total number of TB cases enrolled for the period
With the introduction of automated molecular tests and rapid cultures, it is expected that the proportion of confirmed PTB cases will increase as compared to programmes where only smear microscopy is available.

Proportion of smear-negative PTB
= Number of smear-negative PTB cases enrolled/Total number of TB cases enrolled for the period
This indicator essentially depends on the following: the quality of microscopy, the number of children under treatment (children are rarely smear-positive), the prevalence of HIV infection within the population (these patients present more smear-negative PTB), and the other diagnostics used (culture, Xpert MTB/RIF, etc).
The proportion of smear-negative PTB is about 20% when HIV prevalence is low. It is 40 to 60% when HIV prevalence is high. Proportions that differ significantly from these should make one consider the possibility of under- or over-diagnosis of smear-negative forms.

Proportion of smear-positive PTB
= Number of smear-positive PTB cases enrolled/Total number of TB cases enrolled for the period
In practice, the proportion of smear-positive PTB should correspond to roughly half of all patients. This proportion is lower, however, in areas where HIV prevalence is high. Proportion of smear-positive PTB is around 60% where HIV prevalence is low, and it is 30 to 40% where HIV prevalence is high. Proportions that differ significantly from these should make one consider the possibility of under- or over-diagnosis of smear-negative pulmonary TB and extra-pulmonary B forms.

Proportion of new cases
= Number of new TB cases enrolled/Total number of cases enrolled for the period
This indicator indirectly reflects the relapse and failure rates and possible parallel treatments outside the programme.

Proportion of children
= Number of TB patients less than 15 years enrolled/Total number of TB cases enrolled for the period
Children should represent approximately 10 to 15% of the total number of patients. Proportions that differ significantly from these should make one consider the possibility of under- or over-diagnosis of TB in children.

Proportion of detected cases enrolled under treatment
= Number of cases enrolled under treatment/Total number of cases detected for the period
Patients enrolled are counted from the TB register. Patients detected are counted from the laboratory register and include patients who “interrupted before treatment”.

Case detection rate
= Number of new smear-positive PTB cases detected/Expected number of smear-positive PTB cases for the period
A rough estimate of the expected number of new smear-positive cases can be obtained using the estimated TB incidence given by the WHO in the country profile, which allows an estimate of detection efficacy.

Note: even the best programmes often do not detect more than 60 to 70% of expected new smear-positive cases within a population. In addition, patients might come from outside the target area.

17.4.2 Case detection and enrolment report for DR-TB

See standard DR-TB case detection and enrolment reports in Appendix 33.

Early detection of resistance is intended to ensure that an appropriate treatment is initiated from the start. DST is usually performed for patients at risk of DR-TB. Target groups vary according to local situation, but should at a minimum always include patients who have been previously treated and contacts of confirmed MDR-TB patients.

The indicators for detection aim at measuring the access of TB patients to DST. The frequency of MDR-TB among individuals in different risk groups is also evaluated.

All patients in whom DR-TB is highly suspected or detected should be started on appropriate treatment in the shortest time possible.

A comparison of enrolled patients under treatment to detected DR-TB cases gives an indication of access to care, though some patients started on treatment may have been detected prior to the period of assessment.

The period of assessment is six calendar months. This is usually counted from January to the end of June and July to the end of December. Indicators are measured three months after the end of the six-month period. All data can be extracted from the DR-TB register (Appendix 26), the laboratory register for culture and DST and the Xpert register (Appendix 30).

Each indicator should be calculated for all patients and for each risk group of patients, including: all cases, previously treated cases, failures, household contacts and other local risk groups according to the strategy.

Case detection indicators

Proportion of TB patients detected with DST result for isoniazid and rifampicin (for each risk group during the period)
= Number of TB cases detected with DST result for both isoniazid and rifampicin/Total number of TB cases detected

Proportion of TB patients detected with Xpert MTB/RIF result (for each risk group during the period)
= Number of TB cases detected with Xpert MTB/RIF result/Total number of TB cases detected

Proportion of confirmed MDR-TB cases detected among TB patients tested for isoniazid and rifampicin DST (for each risk group during the period)
= Number of TB cases with confirmed resistance to isoniazid and rifampicin/Total number of TB cases tested for these 2 drugs

Proportion of Xpert RIF resistant cases detected among patients tested by Xpert MTB/RIF (for each risk group during the period)
= Number of Xpert RIF resistant cases/Total number of TB cases with Xpert MTB/RIF result

Enrolment indicators

Proportion of confirmed MDR-TB cases enrolled on MDR-TB treatment
= Number of confirmed MDR-TB cases registered and started on MDR-TB treatment/Total number of confirmed MDR-TB cases detected
This can also be calculated for rifampicin resistant TB cases.

Proportion of confirmed PDR-TB cases enrolled on PDR-TB treatment
= Number of confirmed PDR-TB cases registered and started on PDR-TB treatment/Total number of confirmed PDR-TB cases detected
This calculation does not include rifampicin resistance and unknown isoniazid resistance.

17.4.3 Interim treatment outcomes for drug-susceptible TB and DR-TB

Interim analysis should be completed approximately 3 months after all patients who were registered during a particular interval completed the intensive phase of treatment (three months should allow culture results for all those patients).

Interim treatment outcomes for drug-susceptible TB

Interim results at Month 2 or 3 should be evaluated for all patients treated as new or previously treated patients by standard first-line regimens (with or without confirmation of the drug susceptibility by a DST). These results may be disaggregated by treatment history (new, previously treated, and by type of previous treatment).

At the beginning of a programme, when it is not yet possible to do cohort analysis, the conversion rate at Month 2-3 is a proxy indicator of the effectiveness of treatment, and it allows early detection of potential problems. The smear conversion rate of new smearpositive patients is the proportion of new smear-positive patients who are smear-negative at Month 2. The smear conversion rate of previously treated smear-positive patients is the proportion of previously treated smear-positive patients who are smear-negative at Month 3.

Interim treatment outcomes for DR-TB

The period of assessment is six calendar months, usually counted from January to end June, July to end December. All patients registered and starting treatment during the period of assessment are included in the calculation. The interim report form should be completed 9 months after the closing day of the cohort. This allows culture information at 6 months of treatment to be included for all patients in the cohort. For instance, interim results of TB patients who started treatment during the first semester of a year (1 January to 30 June), should be calculated at the beginning of April of the following year.

Culture conversion (for confirmed DR-TB cases) and death by six months are used as proxies for final outcomes. Information on treatment interruption by six months is helpful. It is also useful to know how many patients started on second-line drugs for MDR-TB turned out not to be MDR.

All data can be extracted from the DR-TB register (Appendix 26).

At six months:

Proportion of death
= Number of confirmed MDR-TB cases registered and started on MDR-TB treatment who died of any cause by the end of Month 6/Total number of confirmed MDR-TB cases started on treatment for MDR-TB during the period

Proportion of treatment interrupted
= Number of confirmed MDR-TB cases started on MDR-TB treatment who interrupted by the end of Month 6/Total number of confirmed MDR-TB cases started on treatment for MDR-TB during the period

Proportion with negative culture
= Number of bacteriologically confirmed pulmonary MDR-TB cases registered and started on MDR-TB treatment with negative culture at Month 6/Total number of bacteriologically confirmed pulmonary MDR-TB cases registered and started on treatment for MDR-TB during the period

Proportion with positive culture
= Number of bacteriologically confirmed pulmonary MDR-TB cases registered and started on MDR-TB treatment with positive culture at Month 6/Total number of bacteriologically confirmed pulmonary MDR-TB cases registered and started on treatment for MDR-TB during the period

Proportion found not to have MDR-TB
= Number of patients started on MDR-TB treatment during the period and later found not to be MDR/Total number of patients started on MDR-TB treatment during the period

17.4.4 Final treatment outcomes for TB

See standard TB treatment outcomes reports (Appendix 32 and Appendix 33).

The final outcome is the most important direct measurement of the effectiveness of a TB programme in terms of patient care. All patients entered on the TB register should be assigned one of six mutually exclusive outcomes at the end of their therapy. All patients should be assigned the first outcome they experience for the treatment being evaluated3.

Final treatment outcome cohort analysis could be carried out when all patients admitted in a given period of time had a chance to complete their treatment. In practice:

– For drug-susceptible TB (and all patients treated by standard first-line regimens) cohort results are analysed quarterly, one year after inclusion of the last patient of the cohort (e.g. cohort of patients admitted during the first quarter 2014 will be evaluated at the end of the first quarter 2015).

– For DR-TB, evaluation occurs 27 months after inclusion of the last patient in the cohort in order to have the results of cultures performed at 24 months. The period of assessment is six calendar months, usually counted from January to the end of June and July to the end of December. All patients starting treatment during this period are included in the calculation. Indicators are measured 24 months after the end of the semester of assessment. All data can be extracted from the DR-TB register.

Although the timing of the analysis is different for drug-susceptible TB and DR-TB, the indicators are the same.

Indicators should be calculated for patients treated by standard first-line regimens (with or without confirmation of drug-susceptible TB by a DST), and for patients with PDR-TB and MDR-TB.
The most important indicators are:

Proportion of cured
= Number of confirmed TB cases declared “cured”/Total number of confirmed TB cases put under treatment during the period
This indicator is calculated for all confirmed drug-susceptible TB cases and DR-TB cases. It is the best indicator of the success of a programme for confirmed TB patients. Though the effectiveness of the treatment for drug-susceptible TB is theoretically above 90%, the proportion of cure is rarely above 70%. For MDR-TB this indicator rarely exceeds 50%.

Proportion of treatment completed
= Number of patients registered as “treatment completed”/Total number of patients put under treatment for the period
A high proportion of patients completing treatment is a positive sign for not confirmed PTB and EPTB. For confirmed TB, it indicates insufficient bacteriological verification at the end of treatment, thus, suggesting that a step should therefore be reinforced.

Proportion with successful outcome
= Number of patients registered as “cured” or “treatment completed”/Total number of patients put under treatment during the period
This is the best indicator to measure the efficacy of a programme for all forms of TB (confirmed and not confirmed, PTB and EPTB). This indicator rarely exceeds 80% for drug-susceptible TB and 60% for MDR-TB.

Proportion of treatment interrupted
= Number of patients registered as “treatment interrupted”/Total number of patients put under treatment during the period
Patients who interrupted treatment are at risk of not being cured or of relapsing. Treatment interruption indicates a failure of the programme in supporting the patient to be able to successfully complete treatment.

Proportion of death
= Number of patients registered as “death”/Total number of patients put under treatment during the period
This ratio usually does not exceed 5% for drug-susceptible TB. Over-mortality may be related to the poor functioning of a programme. It may also be due to a high prevalence of HIV infection among cases or late referrals.

Proportion of failure
= Number of patients registered as “failures”/Total number of patients put under treatment during the period
A high failure rate in new cases can be related to poor treatment adherence, high rate of primary resistance or poor quality of anti-TB drugs. The failure rate should not be over 2% in new cases under treatment.

Proportion of patients for whom HIV status is known
= Number of patients for whom HIV status is known by the end of treatment/Total number of patients put under treatment during the period
This is one of the indicators that help evaluate the integration of TB and HIV services.

TB-HIV co-infection rate
= Number of HIV-infected TB patients/Total number of TB patients put under treatment during the period and for whom HIV status is known at the end of treatment
In high HIV-prevalence regions, co-infection rate may exceed 80%. This information is important in assessing other indicators, in particular the proportion of death.