Tuberculosis
Conditions
Brief summary
The purpose of this study is to compare three strategies for finding TB cases in a rural Sub-Saharan African setting: 1) Screening all attendees of primary care clinics for TB; 2) Conducting household contact investigations of newly diagnosed TB cases; 3) Providing incentives to newly diagnosed TB cases and their contacts to promote contact screening for TB. For each intervention, investigators will measure comparative effectiveness in terms of cases identified as well as the cost-effectiveness and feasibility of implementation.
Interventions
OTHERActive TB case finding
Active TB case finding (ACF) refers to any number of strategies used to identify individuals with active TB disease, outside of passive case finding. In passive case finding, individuals with symptoms present at health centers for diagnosis. In active case finding, the health system makes an effort to identify TB cases before they present passively.
Sponsors
National Institute of Allergy and Infectious Diseases (NIAID)
Perinatal HIV Research Unit of the University of the Witswatersrand
Johns Hopkins Bloomberg School of Public Health
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
DIAGNOSTIC
Masking
NONE
Eligibility
Sex/Gender
ALL
Age
0 Years to 99 Years
Healthy volunteers
Yes
Inclusion criteria
Facility-based screening arm * Age 0-99 years * Informed consent provided (or assent plus parent/guardian consent) * Attending any of the study 28 study clinics in the facility-based screening arm Contact tracing arm- Index Case * Age 0-99 years * Informed consent provided (or assent plus parent/guardian consent) * Newly diagnosed (last 2 months) with TB at any of the 28 study clinics in the contact tracing arm Contact tracing arm- Household Contact * Age 0-99 years * Informed consent provided (or assent plus parent/guardian consent) * Living in the same household as an enrolled Index case (see above) Contact tracing arm- Non-household Close Contact * Age 0-99 years * Informed consent provided (or assent plus parent/guardian consent) * Referred to the study as a close contact of an enrolled Index case (see above)
Exclusion criteria
(all arms): * Unable to provide informed consent.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Treatment Initiation Ratio in Facility Versus Contact Investigation Clinics | 18 months | The primary analysis was based on the facility- level rate ratio, and we first calculated an unadjusted ratio of the treatment initiation rates between the two arms and the corresponding 95% confidence interval (CI). We then adjusted for any residual confounding by district stratification and the historical annual number of people started on TB following a two-stage approach. The first step of this approach fits a Poisson regression to the facility-level counts and the district and historical volume covariates irrespective of study arm. The residuals ratios, calculated as the ratio of the observed over the expected counts, are then used in the second stage to estimate the between-arm rate ratio and the corresponding 95% CI. |
| Comparative Treatment Initiation Ratio in the Incentive-based Versus Household-based Contact Investigation Arms | 36 months | The primary outcome of the study was the comparative number of people with incident TB diagnosed and started on treatment at study clinics in the two contact tracing arms, excluding the six-month washout period. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Estimated Costs Per Contact Person Screened in the Household-based and Incentive-based Contact Investigation Arms | 36 months | These are estimated costs in 2018 USD for the cost per contact person screened in each arm, collected from the societal perspective including health system and patient-level costs. |
| Comparative Number of Secondary TB Cases Identified in Incentive-based Versus Household-based Contact Tracing | 36 months | The pre-specified secondary study outcome was the number of Xpert-based TB diagnoses made among enrolled contacts (secondary cases) by arm. |
| Incremental Cost-effectiveness Ratio | 36 months | (Total costs of incentive-based contact investigation-total costs of household-based contact investigation)/(effectiveness of incentive-based contact investigation-effectiveness of household-based contact investigation) |
| Cost Per Secondary Case of TB Identified Among Contacts, Per Arm | 36 months | Cost were measured in 2018 US dollars from the societal perspective, including both health systems and patient level costs. Total costs per arm were divided by the number of secondary TB cases identified among contacts in that arm. |
| Total Cost of Household Contact Investigation Strategies | 36 months | This measure is the total cost of household contact investigation in both the household-based and incentive-based arms in 2018 United States dollar (USD) collected from the societal perspective (including both health systems and patient-level costs). As these are actual measured (not estimated) costs, no 95% confidence interval is reported. |
Countries
South Africa
Participant flow
Participants by arm
| Arm | Count |
|---|---|
| Facility-based Screening Arm This strategy will be implemented at all clinics (n=28) within this arm for 18 months. Study staff will encourage providers at each of the clinics to screen all consenting patients attending the clinic, regardless of the original reason for clinic presentation. Upon presenting for care (e.g., while waiting for their healthcare provider), patients will be informed about the study and screened for cough of any duration, fever, weight loss, or night sweats. Participants who are symptomatic and provide a sputum specimen (according to the clinic standard of care) will be given a study flyer informing them that they may be contacted by our study staff, and a brief summary of the study. Per standard of care, all sputum samples will be sent to the local National Health Laboratory Service laboratory for Xpert testing. | 1,929 |
| Contact Tracing Household-based Sub-arm This strategy will be implemented for 18 months at half of the clinics within the contact tracing arm (n=14) and, after a six-month washout period, for 18 months in the other half of the clinics in that arm (n = 14). All adults (and guardians for children) with newly diagnosed pulmonary TB will be approached for enrolment as described above. Consenting patients (estimated as 1,575 subjects) will be asked for a home address (with landmarks), mobile phone number, and convenient time for a household visit. A mobile field team consisting of a nurse and lay health worker (both research staff) will circulate between the 14 primary care clinics, with notifications of all new TB cases made from the clinic to the team by mobile phone. The field team will visit the household of each consenting index case within two weeks of notification; each visit will consist of a household census (including demographic data), consent of all eligible household members for questioning and TB screening, administration of a brief questionnaire (including symptomatology, self-reported HIV and TB history, care-seeking behavior, and acceptability of the intervention), and sputum collection for testing with Xpert MTB/RIF. Each participant will also be offered voluntary HIV counselling and testing. The team will make up to three visits to each household as needed in order to approach all household contacts for study enrolment. | 1,413 |
| Contact Tracing Incentive-based Sub-arm This strategy will be implemented in the opposite half of the clinics randomized to contact investigation. All adults (and guardians of children) with newly diagnosed active TB will be identified and approached for enrolment as described above. However, rather than a household visit, each consenting individual will be provided with a set of 10 coupons to give to household members and close contacts whom the index case believes may be at greatest risk. These coupons will contain instructions on how to come to the clinic for free TB screening (e.g., when the study team will be at each clinic), and will have an expiration date of 2 months from the time of initial TB diagnosis. When a contact presents at the clinic with the coupon, the contact will receive a small amount of money and a transport allowance based on the distance travelled to the clinic. If the contact is diagnosed with active TB and starts treatment, the contact will receive an additional larger amount of money. Study staff will document which index case the contact is linked to, and whether the contact is diagnosed with active TB. At the end of the 2 month period in which the coupons are valid the index will receive a grocery voucher for the total incentive earned (e.g., total amount for contact(s) who presented for screening + total amount for contact(s) diagnosed with active TB and starting treatment). | 1,510 |
| Total | 4,852 |
Baseline characteristics
| Characteristic | Facility-based Screening Arm | Contact Tracing Household-based Sub-arm | Contact Tracing Incentive-based Sub-arm | Total |
|---|---|---|---|---|
| Age, Customized Age | 38 years | 39 years | 38 years | 38 years |
| HIV status HIV negative | 562 Participants | 340 Participants | 373 Participants | 1275 Participants |
| HIV status HIV positive | 1161 Participants | 797 Participants | 902 Participants | 2860 Participants |
| HIV status HIV status missing | 1 Participants | 2 Participants | 0 Participants | 3 Participants |
| HIV status HIV status unknown | 205 Participants | 274 Participants | 235 Participants | 714 Participants |
| Race and Ethnicity Not Collected | — | — | — | 0 Participants |
| Region of Enrollment South Africa | 1929 participants | 1413 participants | 1510 participants | 4852 participants |
| Sex: Female, Male Female | 794 Participants | 582 Participants | 630 Participants | 2006 Participants |
| Sex: Female, Male Male | 1135 Participants | 831 Participants | 880 Participants | 2846 Participants |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | 0 / 1,929 | 0 / 1,413 | 0 / 1,510 |
| other Total, other adverse events | 0 / 1,929 | 0 / 1,413 | 0 / 1,510 |
| serious Total, serious adverse events | 0 / 1,929 | 0 / 1,413 | 0 / 1,510 |
Outcome results
Primary
Comparative Treatment Initiation Ratio in the Incentive-based Versus Household-based Contact Investigation Arms
The primary outcome of the study was the comparative number of people with incident TB diagnosed and started on treatment at study clinics in the two contact tracing arms, excluding the six-month washout period.
Time frame: 36 months
Population: This analysis population includes all participants enrolled in the two contact tracing sub-arms during the total study period of 36 months. It excludes participants from the facility-based screening arm. A total of 2,923 patients were started on TB treatment during the study period: 1,413 in clinics allocated to household-based contact investigation and 1,510 in clinics allocated to incentive-based contact investigation.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Facility-based Screening | Comparative Treatment Initiation Ratio in the Incentive-based Versus Household-based Contact Investigation Arms | 1413 Participants |
| Contact Screening | Comparative Treatment Initiation Ratio in the Incentive-based Versus Household-based Contact Investigation Arms | 1510 Participants |
Comparison: The primary outcome of the study was the comparative number of people with incident TB diagnosed and started on treatment at study clinics during the two study periods, excluding the six-month washout period. The number of people starting treatment for incident TB was calculated by performing a review of paper and electronic medical records at each clinic on a quarterly basis during the study.p-value: 0.6895% CI: [0.97, 1.13]Poisson regression
Primary
Treatment Initiation Ratio in Facility Versus Contact Investigation Clinics
The primary analysis was based on the facility- level rate ratio, and we first calculated an unadjusted ratio of the treatment initiation rates between the two arms and the corresponding 95% confidence interval (CI). We then adjusted for any residual confounding by district stratification and the historical annual number of people started on TB following a two-stage approach. The first step of this approach fits a Poisson regression to the facility-level counts and the district and historical volume covariates irrespective of study arm. The residuals ratios, calculated as the ratio of the observed over the expected counts, are then used in the second stage to estimate the between-arm rate ratio and the corresponding 95% CI.
Time frame: 18 months
Population: The population for this outcome consists of a sub-group of the total study population (those enrolled during the intervention period of July 18,2015 - January 17, 2018). All participants from the facility-based screening arm are included (n=1929), and the participants of the contact screening arm (n=1726) are included (this is a combination from the household-based and incentive-based screening arms).
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Facility-based Screening | Treatment Initiation Ratio in Facility Versus Contact Investigation Clinics | 1929 Participants |
| Contact Screening | Treatment Initiation Ratio in Facility Versus Contact Investigation Clinics | 1726 Participants |
Comparison: The primary analysis was based on the facility-level rate ratio. We first calculated an unadjusted ratio of the treatment initiation rates between the two arms and the corresponding 95% CI. We first fit a Poisson regression to the facility-level counts and the district and historical volume covariates. The residuals ratios, calculated as the ratio of the observed over the expected counts, were then used in the second stage to estimate the between-arm rate ratio and the corresponding 95% CI.p-value: 0.7395% CI: [0.8, 1.3]See above comments
Secondary
Comparative Number of Secondary TB Cases Identified in Incentive-based Versus Household-based Contact Tracing
The pre-specified secondary study outcome was the number of Xpert-based TB diagnoses made among enrolled contacts (secondary cases) by arm.
Time frame: 36 months
Population: The number of TB contacts screened in each contact investigation arm.
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Facility-based Screening | Comparative Number of Secondary TB Cases Identified in Incentive-based Versus Household-based Contact Tracing | 17 Participants |
| Contact Screening | Comparative Number of Secondary TB Cases Identified in Incentive-based Versus Household-based Contact Tracing | 17 Participants |
p-value: 0.895% CI: [0.51, 1.95]Poisson regression
Secondary
Cost Per Secondary Case of TB Identified Among Contacts, Per Arm
Cost were measured in 2018 US dollars from the societal perspective, including both health systems and patient level costs. Total costs per arm were divided by the number of secondary TB cases identified among contacts in that arm.
Time frame: 36 months
Population: Individuals included in this population were the total number of secondary TB cases identified among contacts enrolled in each arm.
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Facility-based Screening | Cost Per Secondary Case of TB Identified Among Contacts, Per Arm | 2927 2018 US dollars |
| Contact Screening | Cost Per Secondary Case of TB Identified Among Contacts, Per Arm | 2416 2018 US dollars |
Secondary
Estimated Costs Per Contact Person Screened in the Household-based and Incentive-based Contact Investigation Arms
These are estimated costs in 2018 USD for the cost per contact person screened in each arm, collected from the societal perspective including health system and patient-level costs.
Time frame: 36 months
Population: These populations include all contact persons enrolled into each arm.
| Arm | Measure | Value (MEAN) |
|---|---|---|
| Facility-based Screening | Estimated Costs Per Contact Person Screened in the Household-based and Incentive-based Contact Investigation Arms | 26.44 2018 US dollars |
| Contact Screening | Estimated Costs Per Contact Person Screened in the Household-based and Incentive-based Contact Investigation Arms | 29.93 2018 US dollars |
Secondary
Incremental Cost-effectiveness Ratio
(Total costs of incentive-based contact investigation-total costs of household-based contact investigation)/(effectiveness of incentive-based contact investigation-effectiveness of household-based contact investigation)
Time frame: 36 months
Population: Though the difference in total costs by arm is positive, the difference in effectiveness by arm is zero, therefore we cannot calculate the incremental cost effectiveness ratio as it would require dividing by zero.
Secondary
Total Cost of Household Contact Investigation Strategies
This measure is the total cost of household contact investigation in both the household-based and incentive-based arms in 2018 United States dollar (USD) collected from the societal perspective (including both health systems and patient-level costs). As these are actual measured (not estimated) costs, no 95% confidence interval is reported.
Time frame: 36 months
Population: This population includes all enrolled index participants with TB and their enrolled contacts, per arm.
| Arm | Measure | Value (NUMBER) |
|---|---|---|
| Facility-based Screening | Total Cost of Household Contact Investigation Strategies | 48760 2018 US dollars |
| Contact Screening | Total Cost of Household Contact Investigation Strategies | 58070 2018 US dollars |