The second approach is one where every individual blood pressure measurement
is evaluated with regard to the specific conditions either observed by the researcher
or reported in a diary (often called ecological momentary data) using a variety of
statistical modeling techniques (see for example James et al. 1986 ; Schwartz et al.
1994 ; Kamarck et al. 1998 ; Brondolo et al. 1999 ; Gump et al. 2001 ; Kamarck et al.
2002 , James 2013 ). In these models, blood pressure variation is partitioned based
on the reported diary factors (such as the mood of the subject, the location of the
subject, and the activity at the time). The proportion of variation associated with
each of these aspects is quantified, as is the number of mmHg that alternative
circumstances contribute to, either increasing or decreasing the values of individual
blood pressure measurements relative to some standard value (James2007b, 2013 ).
In evaluating the variation in blood pressure this way, what is reported in the diaries
is critical; that is, the specific characteristics the researcher chooses to have reported
in the diary and the manner in which the factors are recorded (such asfixed
alternatives, open ended responses, or as a ranked scale) will determine how the
variation in blood pressure gets partitioned (James2007b, 2013 ). Analysis of
ecological momentary blood pressure data has been undertaken using raw (e.g.,
Schwartz et al. 1994 ; Brondolo et al. 1999 ; Kamarck et al. 2003 ) and standardized
(e.g., James et al. 1986 ; Ice et al. 2003 ; Brown et al. 2011 ) data. The estimated
effect sizes using these approaches vary considerably, due in part to the fact that
there is no consensus as to what ought to be the standard value against which effects
should be measured, but also because of the demographic and cultural diversity of
the groups studied (James2007a, 2013 ).
Identifying the Magnitude of Hidden Adaptability
How variable is arterial blood pressure over a day, or more specifically, one 24-h
waking–sleep cycle? That is, how much does blood pressure have to change to
adapt us to any given circumstance? This is actually not an easy question to answer.
Figure8.4depicts data from three study subjects whose blood pressure averages
might be termed low, average (or perhaps in today’s medical parlance,
pre-hypertensive), and borderline hypertensive (subject whose resting pressures
range between 140/90 and 160/105 when not taking hypertensive medication).
Specifically, the top panel depicts the circadian variation in a subject whose 24-h
mean pressures were 98/63 mmHg, the middle panel shows the same variation for a
subject with mean pressures 126/80 mmHg, and the bottom panel illustrates the
Fig. 8.4 Circadian blood pressure measurements in subject with low normal (upper panel),c
pre-hypertensive (middle panel), and borderline hypertensive (lower panel) average levels of
blood pressure (98/63, 126/80, and 148/102, respectively) where blood pressures were taken every
15 min from about 8 a.m. to 10 p.m. and every 30 min from 10 p.m. to about 8 a.m. the next
morning. Thex-axis is labeled on a 24-h clock beginning on the day the monitor was applied; thus,
hours labeled 28 and 32 refer to 4 a.m. and 8 a.m. the following day
152 G.D. James