| |
Cook
and Ware (1983)
define a longitudinal study as research in which the same individuals
are observed on more than one occasion. Longitudinal studies can
address the course of pain in individuals already affected, as
well as the development of pain in unaffected individuals. Longitudinal
studies specifically aimed at examining pain onset are classified
as prospective studies (see Section
19). Longitudinal studies are useful for examining individual
change over time.
In the case
of the epidemiologic study of TMD and other pain problems, longitudinal
designs are useful to understand the fluctuating course of pain,
the extent to which pain syndromes remit, recur, or progress,
and to identify prognostic factors predicting future course. Longitudinal
designs are also useful for studying the development of pain syndromes,
as the relationships among pain intensity, activity limitations,
physical findings, and psychological variables can be investigated
over time.
Example
1
|
|---|
| Rammelsberg
et al. (2003) investigated the course of myofascial
TMD pain over a period of 5 years in 235 clinic and
community cases. Subjects were examined at baseline,
and at 1-, 3- and 5-year follow-ups. Subjects meeting
RDC/TMD criteria for myofascial pain at all follow-up
examinations were defined as persistent cases. They
constituted 31% of the sample. Pain remitted for 33%
of subjects and 36% experienced a recurrent course.
Bivariate statistics and multivariate logistic regression
analyses indicated that baseline pain frequency, number
of painful palpation sites and total number of body
sites with pain were significant predictors of becoming
a persistent, as opposed to a remitted or recurrent
case. No predictors could be identified that distinguished
remitted versus recurrent cases. |
|
Example
2
|
|---|
| Following
subjects over time can yield data that further explicates
associations found in cross-sectional studies. For example,
depression is often found to be associated with pain
in cross-sectional studies. Von
Korff and Simon (1996) used longitudinal data to
shed light on this relationship. They found that among
primary care pain patients, depressive symptoms were
initially elevated around the time of the visit, but
then improved to normal levels among patients with a
favorable pain outcome. For patients who continued to
have significant activity limitations, depressive symptoms
remained elevated one year after their visit, but depressive
symptom levels did not increase with time, even if pain
and pain-related activity limitation continued at moderate
to severe levels. These kinds of longitudinal analyses
can begin to shed light on the mechanisms that may produce
associations between pain status and factors that may
be either causes or consequences of pain-related activity
limitation. |
|
|