Rehabilitation is now an integral part of chronic obstructive pulmonary dis
ease (COPD) management. The objective of the study was to determine predict
ors of nonadherence to a COPD rehabilitation programme.
Patients attending a COPD clinic were invited to participate in a 4 week, h
ospital-based, outpatient, COPD rehabilitation programme conducted predomin
antly by respiratory physiotherapists. All potential participants undertook
an interviewer administered questionnaire addressing social, economic, psy
chological and healthcare factors, and underwent baseline physiological mea
sures. Subsequently they were classified as: 1) "adherent" group who comple
ted the total programme (n=55) or 2) "nonadherent" group who refused or beg
an but did not complete the programme (n=36).
The nonadherent group compared to the adherent group were more likely to be
divorced (22 versus 2%, p<0.005), live alone (39 versus 14%, p<0.02), and
to live in rented accommodation (31 versus 6%, p<0.005). There were no diff
erences between the two groups in terms of baseline physiological parameter
s (forced expiratory volume in one second, forced vital capacity, 6-min wal
k distance, oxygen saturation, perceived dyspnoea), quality of life domains
(Chronic Respiratory Disease Questionnaire), or indices of COPD-related mo
rbidity. The nonadherent group were more likely to be current smokers (28 v
ersus 8%, p<0.02) and less Likely to use inhaled corticosteroids (16 versus
42%, p<0.005). The nonadherent group was not significantly likely to be de
pressed, anxious, prone to hyperventilation or to have had previous emotion
al counselling and was more likely to be dissatisfied with disease-specific
social support (51 versus 2%, p<0.001).
In conclusion, a substantial proportion of eligible subjects who did not pa
rticipate in a chronic obstructive pulmonary disease rehabilitation program
me were not more physiologically impaired, but were more likely to be: soci
ally isolated, lack chronic obstructive pulmonary disease-related social su
pport, still be smoking and be less compliant with other healthcare activit
ies. Identification of one or more of these factors reliably allows predict
ion for nonadherence to a rehabilitation programme.