We examined clinical records of all individuals screened for TB and LTBI in two TB outpatient centres from the Northern Health Region: Penafiel (180,000 inhabitants, annual TB incidence rate of 47.7/105 in 2012-16) and Vila Nova de Gaia (330,000 inhabitants, annual TB incidence rate of 26.0/105 in 2012-16).16 Only individuals exposed to an infectious TB patient, diagnosed from 1 January 2015 to 31 December 2016, were included. Exclusion criteria were those described in Figure 1.

Figure 1.

Eligible and included records, with reference to exclusion criteria.

(0.51MB).

We extracted socio-demographic information (sex, age, parish of residence, employment status), clinical information (history of TB/LTBI, immunosuppression, diagnosis of TB during screening), LTBI screening test information (screening strategy, date[s], result[s], diagnosis, initiation of treatment), and information on the index TB cases (diagnosis date, sputum smear results).

Cost-effectiveness was expressed as an incremental cost-effectiveness ratio (ICER), which was calculated by dividing the cost difference between the two screening strategies with the difference number of LTBI diagnosis.11,17 We assumed that treating LTBI is a cost-effective intervention and focused only in the differences between strategies until the moment of LTBI diagnosis.

The median costs in the TST/IGRA and TST-only groups were compared using Mann-Whitney test. The proportion of individuals diagnosed with LTBI, number of visits to the TB outpatient center, and adherence to screening were also compared for these two groups using the Chi-squared test or Fisher's exact test, as appropriate. Adjusted odds-ratios (aORs) and 95% confidence intervals (95% CIs) for each screening strategy and diagnosis of LTBI were calculated using logistic regression. Screening strategy, sex, age, place of residence, professional status (employed/unemployed), TB outpatient center, index case sputum smear positivity, infectious period, and site of disease in the index case were included in the initial regression models. In this procedure, a backward stepwise approach was used, and at each step, the least significant variable that was not a substantial confounder (whose removal would lead to a change of more than 20% in the OR of one or more parameters remaining in the model) was removed. Sex and index case sputum smear positivity were retained regardless of p-value.

Direct individual medical costs were calculated as the sum of estimated costs from the screening test(s), specimen transportation to the testing laboratory, and the work of healthcare professionals (collection and testing of specimens) – data provided by the Northern Regional Health Administration for 2014-2016. IGRA (QuantiFERON-TB Gold Plus) cost was €37.66 (including ELISA kit, antigen, and mitogen) and the TST cost was €1.00 (including tuberculin). Disposable material costs were estimated using online prices (including laboratory materials, blood collection materials, tuberculin needles and syringes, gloves, compresses), and were calculated as €0.31 for one TST and €0.57 for one IGRA.

Direct non-medical costs were estimated per screened individual by multiplying the number of visits to the health center by the distance traveled (calculated taking into account patient's address), with an estimated cost of €0.10 per km. To assess the impact of this estimate on final results, a sensitivity analysis was also performed using an estimated cost of €0.35 per km, the reference value used by the Portuguese Government. This cost was adjusted considering that contacts of the same index case could share their mean of transport if they went to the TB outpatient center on the same day.

Indirect costs per screened individual were calculated by multiplying the number of visits to the health center by the half-daily average income (from Instituto Nacional de Estatística, Statistics Portugal) for working individuals. Half-daily income was used in order to not overestimate indirect costs, considering that TB is associated with socio-economic deprivation.18,19 A sensitivity analysis was also performed using daily average income. The number of potentially averted TB cases was estimated based on the number of individuals who started LTBI treatment, a 10% lifetime risk of developing TB, and a 70% efficacy of treatment (assuming that all patients who began LTBI treatment finished their treatment).20

The following definitions were used in this study:

• •

  Adherence to screening: proportion of individuals who showed up for screening that completed all recommended screening steps;21

• •

  Confirmed respiratory TB patient: person with a positive respiratory TB diagnosis (tracheal, laryngeal, bronchial, pulmonary, and/or pleural), confirmed by a positive culture for Mycobacterium tuberculosis complex (MTC) or a positive smear plus MTC nucleic acid detection in sputum or bronchoalveolar lavage;22

• •

  Close contact: person who had close contact with a patient who had respiratory TB for a cumulative time of at least 8 h if sputum smear-positive, or 40 h if sputum smear-negative (National Tuberculosis Programme recommendation);23

• •

  Immunosuppressed individual: individual receiving chemotherapy, radiotherapy, an immunosuppressive drug, or infected with HIV;21

• •

  Period of infectiousness: time interval during which MTC may be transferred between individuals, estimated as the number of days between onset of symptoms and TB diagnosis;24

• •

  Latent tuberculosis infection: positive IGRA test in an individual who does not have active TB;24

• •

  Medical direct costs: healthcare-related costs (screening tests, specimen transportation to the testing laboratory, and work of healthcare professional);21

• •

  Non-medical direct costs: costs related to the transportation of an individual to a TB outpatient center for screening;21

• •

  Indirect costs: productivity loss from an individual's absence from work because of travel to the TB outpatient center for screening.21

Stata® IC 15.0 (Student version) was used for statistical analysis. Ethical approval for this study was obtained from the Ethics Boards of the Institute of Public Health of the University of Porto (CE17061) and the Northern Regional Health Administration (39/2017).

The median age of screened individuals was 38 years-old (interquartile range [IQR]: 25.0-51.0), 54% were women, and 58% (n = 641) were employed. Their places of residencies were a median of 11 km (IQR 6.5-15.5) and 14 min (IQR 8.5-19.4) away from the visited TB outpatient center. The two groups had no significant differences in terms of age, sex, employment status, and distance to the visited TB center (Table 1). However, the IGRA-only group had a significantly higher proportion of individuals who were exposed to highly infectious TB patients (positive sputum smears) (Table 1). The proportion of contacts diagnosed with LTBI and the adherence to screening were also greater in the IGRA-only group.

After adjusting for sex, age, TB outpatient center, index case sputum smear results, and period of infectiousness, the IGRA-only group had an increased risk for diagnosis of LTBI (aOR = 2.12, 95% CI: 1.48–2.93) (Table 2).

Total average costs were €42.71 per screened individual in the TST/IGRA group and €55.21 in the IGRA-only group; the corresponding median values were €43.71 and €60.23, respectively (Table 3). Medical direct costs were higher in the IGRA group, but non-medical direct costs and indirect costs were higher in the TST/IGRA group (Table 3). The cost per LTBI diagnosis was €280.42 in the TST/IGRA group (76 per 499 screened individuals) and €205.44 in the IGRA group (147 per 547 screened individuals). The estimated number of potentially averted cases of TB was 5 in the TST/IGRA group and 8 in the IGRA-only group. Thus, the cost per potentially averted TB case was €4412.48 in the TST/IGRA group (€21,312.29/4.83) and €3719.20 in the IGRA group (€30,199.87/8.12).

After changing the previous assumptions regarding travel (cost/km of €0.35 instead of €0.10, and average daily income instead of average half-daily income), total costs were €64.48 per screened individual in the TST/IGRA group and €65.63 per individual in the IGRA-only group (median €72.20 and €71.12, respectively) (Table 3). The cost per LTBI diagnosis was €423.36 in the TST/IGRA group (76 in 499 individuals) and €244.22 in the IGRA group (147 in 547 individuals). The cost per potentially averted TB case was €6661.60 in the TST/IGRA group (€32,175.52/4.83) and €4421.13 in the IGRA group (€35,899.61/8.12). Medical direct costs were greater in the IGRA group, but non-medical direct costs and indirect costs were greater in the TST/IGRA group.

The calculated ICER was €106 per LTBI diagnosis, representing increased effectiveness with a slightly increased cost of IGRA-only screening strategy.