Re-examining the impact of malnutrition pre-, during and post-TB treatment through data analysis and modelling of historical and contemporary data - NU/LSHTM project

Title of PhD project / theme

Re-examining the impact of malnutrition pre-, during and post-TB treatment through data analysis and modelling of historical and contemporary data

Supervisory team

Prof Rein Houben: LSHTM (lead)

Prof Sharon Cox: Nagasaki/LSHTM

With additional members of the supervisory team to be identified from LSHTM and Nagasaki according to needs and directions taken by the selected student.

Brief description of project / theme

Although malnutrition (caloric malnutrition measured as body mass index) and weight loss are widely considered important risk factors and symptoms of TB, a lot remains unknown. For example, the exact source, scale and composition of weight loss as a symptom is unknown (GAP 1), even if reasonable evidence is available for low BMI as a risk factor. (Lonnroth et al. 2009).  

In addition, malnutrition receives little attention, compared to other risk factors such as diabetes, smoking, air pollution or HIV in the context of burden estimation. For example, in calculations of TB-associated DALYS (disability-adjusted life years) in GBD, prevalence and distribution of malnutrition are not taken into account. Insufficient data to support causality being given as the reason for this omission.

There is increasing recognition that the impacts of TB extend beyond duration of treatment, and the outcomes of  “cure” or death. Malnutrition, either as a cause of, or caused by the TB episode likely plays a role in these impacts, such as recurrent TB and excess mortality. However, it is rarely considered even in the context of extended household-level poverty and adverse outcomes post-treatment, which may be omitting an important risk factor that can be addressed through intervention.  (GAP 2)

Improved understanding of the continuous role malnutrition plays prior, during and post-TB will begin with a clear framework to map the complex and interrelated relevant determinants, including poverty-related factors associated with TB exposure, such as poor housing, crowding and hygiene and health-related behaviours such as drug and alcohol abuse that may also increase risk of active disease. (GAP 3)

New causal inference statistics, in combination with modelling incorporating new (or rediscovered) concepts in the natural history of TB (Houben et al, 2019) and inclusion of the social & structural determinants of TB and malnutrition and their long-term consequences have the potential to change the current paradigm of malnutrition being cited but not acted upon as central to TB control programmes.

This PhD project would address these gaps through:

GAP 1: A historical literature/data review on the origin and quantitative evidence base for unexpected weight-loss as a symptom of clinical TB disease.

GAP 2: Estimation of TB-associated DALYS including post-TB effects driven by malnutrition.

GAP 3: Develop a quantitative model/framework to explore the multi-faceted contribution of malnutrition (BMI) pre-, during and post-disease to TB-driven loss of health

The project would make use of existing published historical and contemporary data, as well as data from ongoing cohorts led by Prof Cox and others on malnutrition during and after TB treatment.


  • Odone et al. Lancet Diabetes Endocrinol 2014
  • Lonnroth et al. Int J Epi. 2009
  • Pedrazzoli et al. Int J Tuberc Lung Dis 2017.
  • Houben et al. Lancet Respir Med. 2019

The role of LSHTM and NU in this collaborative project

NU are supporting a multidisciplinary programme of TB research in the Philippines lead by Professor Sharon Cox, focussing on the impacts of malnutrition and diabetes in TB patients including long-term post-TB health and wellbeing outcomes. Existing LSHTM-NU collaborations in the programme are supporting two current Joint-PhD students, now in their final year. LSHTM TB center hosts one of the largest multidisciplinary research groups in TB globally, but has less reach in Asia and particularly the Western Pacific region and few people working on malnutrition. This project brings together data from NU-sponsored research with experienced TB modellers to maximise the value of existing and future data collection.

Particular prior educational requirements for a student undertaking this project

Although applicants are not required to have formal qualifications in epidemiology, medical statistics or modelling, applicants must be able to demonstrate good numeracy skills.                                  

To support students in developing the conceptual frameworks of their study questions, the following would be advantageous:

A good grounding in the biomedical sciences for interpretation of animal and human data on potential mechanisms by which malnutrition may be causal in increasing the burden of TB infection and disease.

An appreciation/understanding of social medicine and public health in TB disease.

Skills we expect a student to develop/acquire whilst pursuing this project

Critical review skills of data from diverse sources and ability to synthesise and formulate new hypotheses.

Advanced statistical analysis and modelling