Résumé : Control of tuberculosis has often been managed as a simple issue, the belief being that activities to care for people affected by tuberculosis can be uniformly standardised and centred on the diagnosis and treatment of tuberculosis. The DOTS strategy has been the most concrete illustration of this approach. It is undeniable that this has been successful in re-organising unstructured and very inefficient national TB control programmes.

Today, many countries’ programmes are better organised and have reached case detection and cure rates close to the targets set by WHO (i.e., 70% of cases effectively detected and 85% of detected cases cured). There are mounting arguments to enlarge the scope of activities to care for people affected by TB beyond the classical standardised strategies for diagnostic and treatment of tuberculosis. Indeed, it has become widely accepted that to increase further coverage of diagnosis and treatment of TB, it is necessary to address the economic and psychosocial problems of the people affected by tuberculosis, particularly for those having the least access to and worse quality of care. This will be possible only if, additionally to the current approaches, customised care can be developed after analysis to capture the complexity of care and interventions that take the specificity of local systems in their context into consideration. In chapter 1-2, we illustrate this through the review of the recent customisation of Directly Observed Treatment (DOT) as its naturally evolve in various contexts world-wide.

Developing an analysis that captures complex issues in PATB care means having a proper understanding of the interactions between parts of the local care systems to people affected by TB and identifications of the important patterns of these interactions. That is possible only if information different than the usual quantitative indicators is generated. We illustrate this in the part 2. We took the case of Nicaragua’s TB control programme, which is renowned for its performance in America. In that context, we illustrated the limits of a classical approach to TB control programme evaluation (chapter 2-3) and gave four examples of care process analysis that illustrated the economic and psychosocial problems of people affected by tuberculosis (PATB) (chapter 2-4 to 2-7).

Developing customised system-sensitive interventions to improve the care process means recognising that the interventions cannot be isolated from the organisational context and social dynamics during changes. Thinking must therefore move beyond the design of universal, standardised tool kits. We illustrate specifically in the part 3 the importance of combined local, national and international processes in improving the care process for people affected by TB in Nicaragua: lessons from successful and unsuccessful local and customised processes of implementing interventions in four local health systems (chapter 3-4) can be an opportunity for a health system research unit in a public health school to build a strategic process of care improvement at national level (through scaling up and through the building of a conducive environment) (chapter 3-5).

As a conclusion of this work, we propose in part 4 a three-level reflection through discussion of patterns emerging from the analysis done in the previous chapters: (1) patterns of care and (2) of organisation of health care system are presented in the form of an analytical framework; (3) patterns of regulation and management to improve care for PATB are presented together with a strategy to work on it.