Abstract
Background: To test the hypothesis that there are heterogeneous
patterns of depression development (trajectories) after stroke.
Methods: Data from the South London Stroke Register (1998-2013)
were used. Patients were assessed on socio-demographics, stroke
severity, and medical history at stroke onset, and were followed
up at 3 months after stroke, then annually. Patients were screened
for depression using the Hospital Anxiety and Depression scale
(HADs) at each follow up. A Latent Class Growth Analysis (LCGA)
method based on the Tobit model was used to describe trajectories of
depression over time.
Results: The HAD was completed by 957 (76.4%) of 1247 eligible
patients at 3 months after stroke, and by (80%-87%) at subsequent
assessments. Of patients who survived at least 5 years after stroke
(n=794), four trajectories of depression were identified (Group I-IV),
respectively comprising 16%, 48.5%, 29.1% and 6.4%, Group I,
has no symptoms of depression, mean 1.12 (se 0.37); Group II, and
III have mild and moderate symptoms, mean 3.81(se: 0.29), and
7.5(se:0.34) respectively, and these symptoms increase over time,
Group IV has severe symptoms, mean 13.13 (se:0.30) that persist
over time. The 4 groups differ in many characteristics and Group I,
has male majority (64.6%). Patients with severe physical limitations
comprised 18.1%; 25.7%, 33.3% and 35.3% of group (I-IV) respectively.
Similar groups were identified for longer terms up to 10 years after
stroke.
Conclusions: The LCGA has shown that the development of
depression is not the same for all patients. Four distinctive patterns of
progression were identified providing further insight into the natural
history of depression after stroke.
patterns of depression development (trajectories) after stroke.
Methods: Data from the South London Stroke Register (1998-2013)
were used. Patients were assessed on socio-demographics, stroke
severity, and medical history at stroke onset, and were followed
up at 3 months after stroke, then annually. Patients were screened
for depression using the Hospital Anxiety and Depression scale
(HADs) at each follow up. A Latent Class Growth Analysis (LCGA)
method based on the Tobit model was used to describe trajectories of
depression over time.
Results: The HAD was completed by 957 (76.4%) of 1247 eligible
patients at 3 months after stroke, and by (80%-87%) at subsequent
assessments. Of patients who survived at least 5 years after stroke
(n=794), four trajectories of depression were identified (Group I-IV),
respectively comprising 16%, 48.5%, 29.1% and 6.4%, Group I,
has no symptoms of depression, mean 1.12 (se 0.37); Group II, and
III have mild and moderate symptoms, mean 3.81(se: 0.29), and
7.5(se:0.34) respectively, and these symptoms increase over time,
Group IV has severe symptoms, mean 13.13 (se:0.30) that persist
over time. The 4 groups differ in many characteristics and Group I,
has male majority (64.6%). Patients with severe physical limitations
comprised 18.1%; 25.7%, 33.3% and 35.3% of group (I-IV) respectively.
Similar groups were identified for longer terms up to 10 years after
stroke.
Conclusions: The LCGA has shown that the development of
depression is not the same for all patients. Four distinctive patterns of
progression were identified providing further insight into the natural
history of depression after stroke.
| Original language | English |
|---|---|
| Specialist publication | Clinical Research Design, Rare Diseases and Personalised Medicine |
| Publication status | Accepted/In press - 23 Aug 2015 |