DEVELOPMENT OF MODELS THAT MAY PERMIT EARLY PREDICTION OF
THE RISK OF LONG-TERM SICK LEAVE
Title.
Development of predictive models for the
early identification of patients with greater risk of long
term sick leave due to low back pain.
Background.
Low back pain generates an annual cost equivalent
to 1.7% of the Gross Domestic Product in industrialized countries.
In European countries, more than 90% of these costs are labor
costs, and a minority of patients who represent the most chronic
cases cause the immense majority of the overall costs.
Currently programs are being developed which help reduce
the cost these patients generate. However, these programs
are complex and expensive, so that they cannot be applied
systematically to all cases. Furthermore, they are more effective
when they are applied early, in a phase when it is still not
known which patients are going to become chronic. For that
reason they are used very little in practice.
However, if the factors permitting early prediction (that
is to say, in the first weeks or months of pain) of the factors
that significantly increase the risk of chronification could
be determined, those patients in whom it makes sense to apply
programs focused on reducing the tendency towards chronification
could be identified early. Thus their application would be
more effective and efficient.
Objective.
To develop predictive models that allow the
people with greater risk of becoming chronic patients to be
identified early, with the aim of applying the pertinent programs
without delay.
Design.
Prospective, cohort study.
Methodology.
The study was made in a convenience sample
of 200 patients who could read and write, were actively employed
and had low back pain.
The patients' situation was evaluated on three occasions:
when they were first attended, 14 and 365 days later. On each
occasion, among other variables, all the factors that in international
scientific literature have been said to influence the patient's
occupational prognosis were collected, as well as the clinical
variables, such as the exact diagnosis, clinical history,
pain intensity and degree of disability-, variables indicative
of the treatment strategy applied by the doctors, such as
diagnostic tests, treatments and prescribed referrals, and
variables of a psycho-social kind, such as the patient's beliefs
in regard to his or her back pain, overall quality of life,
activity and occupational situation or educational level.
Finally, the occupational evolution of each patient during
the period of the study was gathered, determining the number
of sick leave days due to low back pain. In the analysis phase,
models were established that, combining the information collected
in the first two evaluations, predicted the occupational evolution
over the follow-up year, and identified the factors without
or with prognostic value, specifying in this last case, the
meaning and magnitude of their influence.
Participants, along with the Foundation's
Science Department.
Unit of Clinical Biostatistics at the Hospital
Ramón y Cajal, Madrid; Provincial Delegation of Social
Services, Granada; Departments of Neurosurgery and Traumatology
at the Hospital Universitario de Son Dureta, Palma de Mallorca;
Department of Preventative Medicine, Hospital Virgen de las
Nieves, Granada; 14 Primary Care Health Centers in 6 Autonomous
Communities (Regions of Spain); Spinal Surgery Unit at the
Valld'Hebrón, Barcelona; Department of Rheumatology
at the Hospital Clinic, Barcelona; Mutua Asepeyo, Madrid;
Physiotherapy Unit, Pain Clinic and Spine Unit at the Hospital
de Sabadell (Corporación Sanitaria Parc Taulí),
Barcelona; Department of Psychology at the Universidad Autónoma,
Bellaterra, Barcelona; Department of Traumatology at the Fundación
Jimenez Díaz, Madrid; Advanced Research Techniques
in Health Services (TAISS), Madrid; who are members of the
Spanish Network of Researchers in Back Disorders.
The study was funded in its entirety by the Kovacs Foundation.
Status.
In progress.
|
Home
