582 Panel Data Methods
that measures of body fat are sometimes used instead of BMI. The height-for-age
z-score standardizes a child’s measured height using the median (or mean) and
standard deviation for children of the same age and sex from a reference popu-
lation, such as the US National Center for Health Statistics reference population
of well-nourished American children (e.g., Duflo, 2000). Height is also compared
to a standard distribution to construct measures of stunting (e.g., Gertler, 2004).
For example, Chen and Zhou (2007) use height to measure the long-term health
consequences of childhood exposure to the 1959–61 famine in China, which is
estimated to have caused 15–30 million excess deaths. They adopt a difference-
in-differences approach that exploits regional differences in exposure to the
famine.
Anthropometric measures may also play a role as biomarkers (discussed in more
detail below). For example, height and weight can be used as predictors of mortal-
ity, stroke and cardiovascular disease and the waist-to-hip ratio is a predictor for
hypertension, late-onset diabetes, cardiovascular disease, stroke and some forms
of cancer.
12.3.2.3 Biomarkers
Biological markers, or biomarkers, are likely to play an increasing role in future
research in health economics as they become incorporated into an increasing range
of datasets, including longitudinal datasets, such as the US Health and Retirement
Survey (HRS), the English Longitudinal Survey of Ageing (ELSA) and the planned
UK Longitudinal Household Study. This trend is likely to be enhanced by the avail-
ability of DNA information and genetic screening, which provide greater potential
to control for individual heterogeneity. Biomarkers are biological or physiological
measures that indicate the presence of a disease or the propensity to develop a dis-
ease. They can be used to identify risk factors and as objective measures of health
that avoid contamination by reporting bias (see, e.g., Adda and Cornaglia, 2006;
Bankset al., 2006; Currieet al.,2007).
Biomarkers for cardiovascular disease include elevated blood pressure and vari-
ability in the heart rate. Metabolic biomarkers include serum HDL (high-density
lipoprotein) and total cholesterol and triglycerides, which are predictors of heart
disease; fibrinogen, which is linked to blood clotting and the risk of heart disease;
and glycated haemoglobin, which is a proxy indicator for diabetes. Biomarkers
linked to the immune system include interleukin-6 (IL-6), which is a predic-
tor of Alzheimer’s disease, arthritis, diabetes and osteoporosis; C-reactive protein
(CRP), which indicates lupus, pneumonia, rheumatoid arthritis, rheumatic fever
and tuberculosis; ferritin and hemoglobin, which indicate iron deficiency; serum
retinol, which indicates vitamin A deficiency. Biomarkers linked to hormonal indi-
cators of stress (HPA axis) include cortisol, adrenocorticotropic hormone (ACTH)
and dehydroepiandrosterone-sulphate (DHEA-S). Biomarkers linked to the sympa-
thetic nervous system include norepiphrenine, which is associated with longevity,
and epiphrenine (adrenaline), which is linked to cognitive decline and longevity.
Biomarkers may also be used as objective indicators of physical functioning and