Isoniazid preventive therapy completion in children under 5 years old who are contacts of tuberculosis cases in Lima, Peru: study protocol for an open-label, cluster-randomized superiority trial.

BACKGROUND: Children < 5 years old in contact with TB cases are at high risk for developing severe and fatal forms of TB. Contact investigation, BCG vaccination, and isoniazid preventive therapy (IPT) are the most effective strategies to prevent TB among children. However, the implementation of IPT faces challenges at several stages of the cascade of care of TB infection among children, particularly those less than 5 years old. In Peru, a large proportion of children do not complete IPT, which highlights the need to design effective interventions that enhance preventive therapy adherence and completion. Although the body of evidence for such interventions has grown, interventions in medium TB incidence settings are lacking. This study aims to test the effectiveness, acceptability, and feasibility of an intervention package to increase information and motivation to complete IPT among children < 5 who have been prescribed IPT. METHODS: An open-label, cluster-randomized superiority trial will be conducted in two districts in South Lima, Peru. Thirty health facilities will be randomized as clusters, 10 to the intervention and 20 to control (standard of care). We aim to recruit 10 children from different households in each cluster. Participants will be caretakers of children aged < 5 years old who initiated IPT. The intervention consists of educational material, and short message services (SMS) reminders and motivators. The primary outcomes will be the proportion of children who picked up > 90% of the 24 weeks of IPT (22 pick-ups) and the proportion of children who picked up the 24 weeks of IPT. The standard of care is a weekly pick-up with monthly check-ups in a health facility. Feasibility and acceptability of the intervention will be assessed through an interview with the caretaker. DISCUSSION: Unfavorable outcomes of TB in young children, high effectiveness of IPT, and low rates of IPT completion highlight the need to enhance adherence and completion of IPT among children < 5 years old. Testing of a context-adapted intervention is needed to improve IPT completion rates and therefore TB prevention in young children. TRIAL REGISTRATION: ClinicalTrials.gov NCT03881228. Registered on March 19, 2019.

Real-time parameter estimation of Zika outbreaks using model averaging.

Early prediction of the final size of any epidemic and in particular for Zika disease outbreaks can be useful for health authorities in order to plan the response to the outbreak. The Richards model is often been used to estimate epidemiological parameters for arboviral diseases based on the reported cumulative cases in single- and multi-wave outbreaks. However, other non-linear models can also fit the data as well. Typically, one follows the so called post selection estimation procedure, i.e., selects the best fitting model out of the set of candidate models and ignores the model uncertainty in both estimation and inference since these procedures are based on a single model. In this paper we focus on the estimation of the final size and the turning point of the epidemic and conduct a real-time prediction for the final size of the outbreak using several non-linear models in which these parameters are estimated via model averaging. The proposed method is applied to Zika outbreak data in four cities from Colombia, during the outbreak ocurred in 2015-2016.

Time to initiation of multidrug-resistant tuberculosis treatment and its relation with outcome in a high incidence district in Lima, Peru.

OBJECTIVE: To determine the time from diagnosis to start of multidrug resistant tuberculosis (MDR TB) treatment in Lima, Peru. METHODS: We studied new smear-positive TB adults that were started on MDR TB treatment or that were switched to it between June 2008 and December 2011. RESULTS: Time from the first positive smear to MDR-TB treatment was >30 days in 35% (13/37) of patients. Among the 27% (24/88) of patients that switched to MDR-TB treatment, time from the last dose of a drug-susceptible regimen was >30 days. CONCLUSION: Start of and switching to MDR TB treatment is still delayed.

Manual liquid culture on simple Middlebrook 7H9 or MGIT for the diagnosis of smear-negative pulmonary tuberculosis.

OBJECTIVES: To compare the performance of liquid culture on simple Middlebrook 7H9 to the one of manual mycobacterial growth indicator tube (MGIT) and solid culture on Ogawa for the diagnosis of smear-negative tuberculosis (SN-TB) in a high-burden, resource-constrained setting. METHODS: Sputum samples from patients with clinical suspicion of SN-PTB admitted to two-third-level hospitals in Lima between September 2005 and May 2008 were cultured in parallel on simple Middlebrook 7H9, manual MGIT and Ogawa. A case of SN-TB was defined as one with a positive culture in any medium. RESULTS: Among samples from 542 patients, 151 (28%) cases of SN-TB were identified. The sensitivity of Middlebrook 7H9 (0.76, 95% CI 0.69-0.83) was not substantially different from that of MGIT (0.85, 95% CI 0.79-0.91). Ogawa had the lowest sensitivity (0.63, 95% CI 0.55-0.71). The median turnaround time was similar for both liquid media (18 days), and it was shorter than that of Ogawa (30 days). CONCLUSIONS: Culture on simple Middlebrook 7H9 performs almost as well as MGIT, at a probably more affordable cost. Further studies on the cost-effectiveness of this overlooked technique should be performed.

Quality assessment of smear microscopy by stratified lot sampling of treatment follow-up slides.

SETTING: Ten peripheral laboratories performing routine acid-fast bacilli (AFB) smear microscopy in Lima, Peru. OBJECTIVES: To test whether external quality assessment (EQA) rechecking of AFB smears becomes more efficient with stratified lot sampling of treatment follow-up smears. DESIGN: In 2 consecutive years, a stratified lot sample of 36 treatment follow-up slides and 24 diagnostic slides were randomly selected and blindly rechecked. A second controller determined the final result for discordant slides. Feedback was provided to laboratory technicians during supervisory visits. RESULTS: More false-negative errors were found in the follow-up slides than in the tuberculosis suspect slides: 25 vs. 3. This represented a yield of 3.5% in 720 follow-up slides and only 0.6% in 480 diagnostic slides. Positive predictive values were high in both years. Respectively three and eight laboratories did not reach a relative sensitivity of >65% during the first and second year, and a clear improvement was seen in only one laboratory. Excessive workload seemed to preclude raising the level of routine performance. CONCLUSIONS: EQA with stratified lot sampling of treatment follow-up slides proved very efficient and effective for identifying laboratories with substandard performance in a setting with low positivity rates in routine diagnostic smears.

Determinants of uptake, short-term and continued use of insecticide-treated curtains and jar covers for dengue control.

OBJECTIVE: To evaluate the acceptance and long-term use of insecticide-treated (IT) materials for dengue vector control. METHODS: In 2007, IT jar covers and/or curtains (PermaNet®) were distributed under routine conditions to 4101 households (10 clusters) in Venezuela and to 2032 households (22 clusters) in Thailand. The use of IT tools was measured at distribution (uptake), at 5/6 months (short-term use) and at 18/22 months (continued use) after distribution. Determinants of use were assessed with logistic regression analysis. RESULTS: The uptake of IT curtains was 76.7% in Venezuela and 92.3% in Thailand. It was associated with being a resident for >5 years (OR Venezuela 3.0 95% CI 2.0-4.4; OR Thailand 3.5 95% CI 1.7-7.3) and with pre-intervention use of ordinary curtains (OR Venezuela 2.2 95% CI 1.4-3.6). The continued use decreased significantly to 38.4% of households in Venezuela and 59.7% in Thailand and was, conditional on short-term use, only determined by the perceived effectiveness of IT curtains (OR Venezuela 13.0 95%CI 8.7-19.5; OR Thailand 4.9 95% CI 3.1-7.8). Disease knowledge and pre-intervention perception of mosquito nuisance were not associated with IT curtains' uptake or use. The uptake of IT jar covers in Venezuela was 21.5% and essentially determined by the presence of uncovered jars in the household (OR 32.5 95% CI 14.5-72.6). Their continued use, conditional on short-time use, was positively associated with the household use of Abate® (OR 7.8 95% CI 2.1-28.9). CONCLUSION: The use of IT curtains rapidly declines over time. Continued use is mainly determined by the perceived effectiveness of the tool. This poses a real challenge if IT curtains are to be introduced in dengue control programmes.

Duration of cough, TB suspects' characteristics and service factors determine the yield of smear microscopy.

OBJECTIVE: To determine the efficiency of routine tuberculosis (TB) case detection by examining sputum smear positivity for acid-fast bacilli in relation to duration of cough, characteristics of TB suspects examined and health service factors. METHOD: We combined patient interviews with routine data from laboratory registers in 6 health care facilities in San Juan de Lurigancho district, Lima, Peru. A TB case was defined as a TB suspect with at least one positive sputum smear. We calculated adjusted odds ratios with 95% confidence intervals for the association between smear positivity and health service and patient's characteristics. RESULTS: Smear positivity was 7.3% (321/4376). Of the 4376 adults submitting sputa, 55.3% (2418) reported cough for <14 days. In this group, smear microscopy yielded 3.2% (78/2418) positive results vs. 12.4% (243/1958) in patients coughing for 14 or more days. Having cough for >2 weeks, being referred by health care staff, attending a secondary-level health care facility, male sex and age between 15 and 44 years were independent determinants of smear positivity. CONCLUSIONS: Routine case detection yields a low proportion of smear-positive cases because of the inclusion of a high proportion of patients without cough or coughing for <2 weeks. Adherence to the national TB control programme guidelines on the selection of TB suspects would have a positive impact on the smear positivity rate, reduce laboratory costs and workload and possibly improve the reading quality of smear microscopy.

Breteau Index threshold levels indicating risk for dengue transmission in areas with low Aedes infestation.

We used Breteau Indices to - retrospectively - predict dengue transmission. The presence of one house block (on average about 50 houses) with a Breteau Index >or=4 in a neighbourhood (a block plus the surrounding blocks in a radius of 100 m) predicted transmission at the latter level with 81.8% sensitivity and 73.3% specificity during the first month of the Havana 2001 dengue outbreak. This result corroborates the external validity of this threshold, at least in situations with low infestations levels.

Community involvement in dengue vector control: cluster randomised trial.

OBJECTIVE: To assess the effectiveness of an integrated community based environmental management strategy to control Aedes aegypti, the vector of dengue, compared with a routine strategy. Design Cluster randomised trial. Setting Guantanamo, Cuba. Participants 32 circumscriptions (around 2000 inhabitants each). Interventions The circumscriptions were randomly allocated to control clusters (n=16) comprising routine Aedes control programme (entomological surveillance, source reduction, selective adulticiding, and health education) and to intervention clusters (n=16) comprising the routine Aedes control programme combined with a community based environmental management approach. MAIN OUTCOME MEASURES: The primary outcome was levels of Aedes infestation: house index (number of houses positive for at least one container with immature stages of Ae aegypti per 100 inspected houses), Breteau index (number of containers positive for immature stages of Ae aegypti per 100 inspected houses), and the pupae per inhabitant statistic (number of Ae aegypti pupae per inhabitant). RESULTS: All clusters were subjected to the intended intervention; all completed the study protocol up to February 2006 and all were included in the analysis. At baseline the Aedes infestation levels were comparable between intervention and control clusters: house index 0.25% v 0.20%, pupae per inhabitant 0.44 x 10(-3) v 0.29 x 10(-3). At the end of the intervention these indices were significantly lower in the intervention clusters: rate ratio for house indices 0.49 (95% confidence interval 0.27 to 0.88) and rate ratio for pupae per inhabitant 0.27 (0.09 to 0.76). CONCLUSION: A community based environmental management embedded in a routine control programme was effective at reducing levels of Aedes infestation. Trial Registration Current Controlled Trials ISRCTN88405796.

Intersectoral coordination, community empowerment and dengue prevention: six years of controlled interventions in Playa Municipality, Havana, Cuba.

OBJECTIVE: To document the process, outcome and effectiveness of a community-based intervention for dengue control. METHODS: The primary intervention, focused on strengthening intersectoral coordination, was initiated by researchers in January 2000 in a pilot area in Playa municipality, Havana. In August 2002 health authorities extended the intervention to neighbouring areas, one of which was selected for evaluation. In August 2003 a complementary strategy, focused on community empowerment, was initiated in half of the pilot area. In our control area, routine dengue activities continued throughout the study period. Longitudinal process assessment was carried out using document analysis, interviews and group discussions. Random population surveys in 1999, 2002 and 2005 assessed levels of participation and behavioural changes. Entomological surveillance data from 1999 to 2005 were used to determine effectiveness. RESULTS: Mean scores for participation in the pilot area were 1.6, 3.4 and 4.4 at baseline, and 2 years after initiating intersectoral coordination and intersectoral coordination plus community empowerment interventions, respectively. While in the control area little behavioural change was observed over time, changes were considerable in the pilot and extension areas, with 80% of households involved in the community empowerment intervention showed adequate behavioural patterns. The pilot and extension areas attained comparable entomological effectiveness with significantly lower Breteau indices (BIs) than the control area. The pilot (sub-) area with the community empowerment intervention reached BIs below 0.1 that continued to be significantly lower than the one in the control area until the end of the study. CONCLUSION: The study showed a trend in the levels and quality of participation, behavioural change and effectiveness of Aedes control from the routine activities only over an intervention with intersectoral coordination to one that combined intersectoral coordination and community empowerment approach.

Community involvement in dengue vector control: cluster randomised trial.

OBJECTIVE: To assess the effectiveness of an integrated community based environmental management strategy to control Aedes aegypti, the vector of dengue, compared with a routine strategy. DESIGN: Cluster randomised trial. SETTING: Guantanamo, Cuba. PARTICIPANTS: 32 circumscriptions (around 2000 inhabitants each). INTERVENTIONS: The circumscriptions were randomly allocated to control clusters (n=16) comprising routine Aedes control programme (entomological surveillance, source reduction, selective adulticiding, and health education) and to intervention clusters (n=16) comprising the routine Aedes control programme combined with a community based environmental management approach. MAIN OUTCOME MEASURES: The primary outcome was levels of Aedes infestation: house index (number of houses positive for at least one container with immature stages of Ae aegypti per 100 inspected houses), Breteau index (number of containers positive for immature stages of Ae aegypti per 100 inspected houses), and the pupae per inhabitant statistic (number of Ae aegypti pupae per inhabitant). RESULTS: All clusters were subjected to the intended intervention; all completed the study protocol up to February 2006 and all were included in the analysis. At baseline the Aedes infestation levels were comparable between intervention and control clusters: house index 0.25% v 0.20%, pupae per inhabitant 0.44x10(-3) v 0.29x10(-3). At the end of the intervention these indices were significantly lower in the intervention clusters: rate ratio for house indices 0.49 (95% confidence interval 0.27 to 0.88) and rate ratio for pupae per inhabitant 0.27 (0.09 to 0.76). CONCLUSION: A community based environmental management embedded in a routine control programme was effective at reducing levels of Aedes infestation. TRIAL REGISTRATION: Current Controlled Trials ISRCTN88405796.

Diagnostic validity of an expert tuberculosis commission that assists the diagnosis of bacteriologically negative suspected TB cases in Havana, Cuba.

The Provincial Tuberculosis Commission of Havana, Cuba, a multi-speciality committee, assists clinicians in diagnosing bacteriologically negative tuberculosis (TB). At its weekly meetings, clinicians present the files of suspected TB cases for discussion, diagnosis and recommendations. This prospective study assessed the validity of the diagnoses made by the Commission by comparing the diagnoses made with diagnoses ascertained after one year of follow-up. Between October 2002 and December 2003, 126 patients suspected to have TB but who were bacteriologically negative completed diagnostic work at the Commission. Fifty-three (42%) were diagnosed as TB cases. The definite diagnosis of 116 patients (92%) was ascertained after one year of follow-up. Six patients diagnosed by the Commission as TB cases were suffering from other diseases, while one patient diagnosed with pneumonia had a definite diagnosis of pulmonary TB. The diagnostic sensitivity and specificity of the Commission were 98% (95% CI 93-100) and 92% (95% CI 85-98), respectively. The Provincial Tuberculosis Commission of Havana can be considered a valuable tool for the diagnosis of TB in patients suspected of TB but who are bacteriologically negative. A comparable approach, adapted to the local conditions, could prove useful in other epidemiological and healthcare settings.

A clinical prediction rule for pulmonary tuberculosis in emergency departments.

SETTING: University-affiliated hospital located in an area with a high incidence of pulmonary tuberculosis (PTB). OBJECTIVE: To develop a clinical prediction rule (CPR) based on information obtainable on admission, to permit rapid identification of patients with PTB. DESIGN: Information from patients with respiratory symptoms who attended the emergency department of Cayetano Heredia Hospital, Lima, Peru, was collected prospectively. Clinical symptoms, past medical history, demographic data and results of chest X-rays (CXRs), sputum smear and culture in Löwenstein-Jensen media were obtained. Based on logistic regression, we constructed a scoring system to predict PTB. RESULTS: A total of 345 patients were enrolled in the study, including 109 (31%) culture-proven PTB cases. In logistic regression analysis, we found age, previous history of PTB, weight loss, presence of cavities, upper lobe infiltrate and miliary pattern on CXR as independent predictors of PTB. We designed a scoring system with these variables, taking into account their statistical weight. The score attained 93% sensitivity and 42% specificity. CONCLUSION: The CPR that was developed performed well in our population. It merits further validation in other settings. It should not, however, replace, but should complement sputum microscopy when deciding on isolation, and it does not preclude microbiology in making a definitive diagnosis.

Could clinical audit improve the diagnosis of pulmonary tuberculosis in Cuba, Peru and Bolivia?

OBJECTIVES: To assess the effectiveness of clinical audit in improving the quality of diagnostic care provided to patients suspected of tuberculosis; and to understand the contextual factors which impede or facilitate its success. METHODS: Twenty-six health centres in Cuba, Peru and Bolivia were recruited. Clinical audit was introduced to improve the diagnostic care for patients attending with suspected TB. Standards were based on the WHO and TB programme guidelines relating to the appropriate use of microscopy, culture and radiological investigations. At least two audit cycles were completed over 2 years. Improvement was determined by comparing the performance between two six-month periods pre- and post-intervention. Qualitative methods were used to ascertain facilitating and limiting contextual factors influencing change among healthcare professionals' clinical behaviour after the introduction of clinical audit. RESULTS: We found a significant improvement in 11 of 13 criteria in Cuba, in 2 of 6 criteria in Bolivia and in 2 of 5 criteria in Peru. Twelve out of 24 of the audit criteria in all three countries reached the agreed standards. Barriers to quality improvement included conflicting objectives for clinicians and TB programmes, poor coordination within the health system and patients' attitudes towards illness. CONCLUSIONS: Clinical audit may drive improvements in the quality of clinical care in resource-poor settings. It is likely to be more effective if integrated within and supported by the local TB programmes. We recommend developing and evaluating an integrated model of quality improvement including clinical audit.

Improving sputum microscopy services for the diagnosis of tuberculosis in Peru and Bolivia.

SETTING: Sixteen primary care health centres in Peru and Bolivia. OBJECTIVES: To assess the utilisation of microscopy services in Peru and Bolivia and determine if clinical audit, a quality improvement tool, improves the utilisation of these services. DESIGN: We estimated the percentage of patients with suspected tuberculosis (TB) in whom sputum microscopy was effectively utilised in Peru and Bolivia over two 6-month periods before and after a clinical audit intervention that included standards setting, measuring clinical performance and feedback. RESULTS: Before the intervention, only 31% (95%CI 27-35) of TB suspects were assessed with sputum microscopy in Peru. In Bolivia, 30% (95%CI 25-35) underwent at least two sputum microscopy examinations. After clinical audit, the availability of sputum microscopy results improved by respectively 7% (95%CI 1-12, P < 0.05) and 23% (95%CI 15-30, P < 0.05) over 2 years in Peru and Bolivia. CONCLUSIONS: Despite World Health Organization recommendations that all TB suspects should undergo sputum microscopy before treatment, results are available for further assessment for only one third. This is a potentially serious obstacle to TB case detection. Clinical audit can bring some improvement. We recommend regular monitoring of effective utilisation of microscopy services and investigations to ascertain organisational and structural issues in their uptake and use.

Community participation in Aedes aegypti control: a sociological perspective on five years of research in the health area ''26 de Julio'', Havana, Cuba.

OBJECTIVE: Effective dengue prevention and Aedes aegypti control is a priority for the Cuban health authorities. To enhance effectiveness, strategies oriented towards a more active involvement of communities in control activities are being tested. This paper presents a sociological perspective on a pilot project conducted in the health area ''26 de Julio'' (La Havana) in 1999-2004. METHODS: Instrumental case study based on an exhaustive content analysis of project documents and on observations of a sociologist. RESULTS: The context and the pilot project are systematically described and an analysis of the evolution of the underlying concept of community participation is provided. The pilot experience was a dynamic process influenced by self-reflection of the research team, feedback from research partners and changes in the epidemiological context (provoked by two dengue outbreaks during the study period). Community participation evolved from being just one component in Aedes aegypti control directed by the health staff into a learning and empowering process for the people. This change in the concept of participation was reflected in different aspects of the pilot project such as the learning and evaluation processes. CONCLUSION: Empirical evidence from 5 years of research in the particular context of Cuba showed that moves towards community-based Aedes aegypti control are feasible. However, in order to be successful, community-based dengue prevention should be a social learning process, implying a transfer of power and responsibilities to local people. Actions undertaken must be oriented towards creating local capabilities, strengthening existing structures and organizations and promoting group work for learning participation from participation itself.

Cost effectiveness of Aedes aegypti control programmes: participatory versus vertical.

We conducted an economic appraisal of two strategies for Aedes aegypti control: a vertical versus a community-based approach. Costs were calculated for the period 2000-2002 in three pilot areas of Santiago de Cuba where a community intervention was implemented and compared with three control areas with routine vertical programme activities. Reduction in A. aegypti foci was chosen as the measure of effectiveness. The pre-intervention number of foci (614 vs. 632) and economical costs for vector control (US$243746 vs. US$263486) were comparable in the intervention and control areas. During the intervention period (2001-2002), a 13% decrease in recurrent costs for the health system was observed. Within the control areas, these recurrent relative costs remained stable. The number of A. aegypti foci in the pilot areas and the control areas fell by 459 and 467, respectively. The community-based approach was more cost effective from a health system perspective (US$964 vs. US$1406 per focus) as well as from society perspective (US$1508 vs. US$1767 per focus).

Towards active community participation in dengue vector control: results from action research in Santiago de Cuba, Cuba.

Community participation is advocated as essential for attaining effective dengue prevention, but knowledge of how to foster this is limited. In Santiago de Cuba, multiple small task forces were created at the neighbourhood level that included all stakeholders in the control of Aedes aegypti. The task forces assessed the perceived needs and elaborated action plans to promote specific behavioural change and to reduce environmental risks through social communication strategies and intersectoral local government activities. We monitored five dimensions of the participation process and assessed behavioural and environmental results and entomological outcomes. Participation was weak to good. At the household level, uncovered water storage containers decreased from 49.3% to 2.6% between 2000 and 2002, and removing larvicide from them dropped from 45.5% to 1%. There was a reduction of 75% in the absolute number of positive containers and a significant decrease from 1.23% to 0.35% in the house index. Local task forces, in which the interests of householders as well as vector control workers are directly represented, can lead to effective government-community partnerships that resolve problems of mutual concern.

Delays to treatment and out-of-pocket medical expenditure for tuberculosis patients, in an urban area of South America.

Short delays to treatment are important for the control of tuberculosis (TB). National Tuberculosis Programmes provide free diagnosis and treatment for smear-positive patients, so that the patients' out-of-pocket medical expenditure could be almost nil. The factors associated with delays in starting treatment, and the pre-treatment out-of-pocket medical expenditure for TB patients, have now been investigated in the Bolivian city of Cochabamba. Bolivia is the Latin American country with the highest incidence of TB. It is covered by a national TB programme that provides free diagnosis and free treatment for smear-positive patients. Structured interviews with 144 smear-positive patients enrolled in this programme revealed median patient, provider and total delays of 3.6, 6.2 and 12.9 weeks, respectively. The total delays were longer for the female patients than for the male, and for patients who consulted private doctors than for the other patients. When the first healthcare provider was a doctor, the median provider delay was 4.9 weeks in the public sector but 7.2 weeks in the private. The median out-of-pocket medical expenditure per patient, which was U.S.$13.2 overall, was much higher for those who consulted a private doctor than for those who did not (U.S.$21.9 v. U.S.$5.4, respectively; P<0.001). It appears that interventions targeting doctors (in both the private and public sectors) are likely to have a larger impact on the shortening of delays in TB treatment than interventions targeting patients. They could also reduce unnecessary out-of-pocket expenditure.

Collaboration between private pharmacies and national tuberculosis programme: an intervention in Bolivia.

BACKGROUND: Public-private partnerships are felt to be necessary for tuberculosis (TB) control in some developing countries. OBJECTIVES: To evaluate the potential of a collaboration between the National TB Programme (NTP) and private pharmacies in Bolivia, the country with the highest TB incidence in Latin America. METHODS: We contacted the local Pharmacists' Association in the city of Cochabamba, and designed a two phase intervention. The objectives of the first phase were to decrease the availability of TB drugs in private pharmacies on a voluntary basis, and to improve referral of clients seeking TB drugs to the NTP. A survey of all pharmacies allowed for a before-after comparison with a baseline survey. The objectives of the second phase were to obtain referral of pharmacy clients with chronic cough for TB screening in the NTP. This phase was started in 70 pharmacies and evaluated after 2 months using the referral slips issued by the pharmacists. RESULTS: The proportion of pharmacies selling TB drugs decreased (rifampicin: 23-11.5%; isoniazid: 16-3.1%; P<0.001) and the proportion of pharmacies referring to the NTP clients seeking TB drugs increased (22-58%; P<0.0001). In the second phase, 26 of 70 pharmacies (38%) referred a total of 41 clients for screening in the NTP (i.e. an average of 0.29 clients per pharmacy and per month); 11 of 41 (27%) were screened and three of 11 (27%) diagnosed with smear-positive TB. CONCLUSION: The first phase of the intervention proved effective in reducing the availability of the main TB drugs in pharmacies, and in improving referral of clients seeking TB drugs. Key factors in this success were not specific to Bolivia, and collaboration between private pharmacies and public services appears possible in that respect. However, collaboration with pharmacies does not seem an efficient way to increase the number of patients screened for TB, and to shorten delays to TB diagnosis and treatment.