The proposed study will examine, for the first time, the 12-month course of apathy and its impact on functional outcomes in the largest cohort of SAH survivors to date. The first hypothesis is that 42% (51) of patients with apathy at baseline (3 month post-SAH, T1) will continue to exhibit apathy at 12 months after the first assessment. The second hypothesis is that poor cognitive status, a high level of comorbidity, and the presence of depressive symptoms at baseline (27–29) are predictors of persistent apathy. The third hypothesis is that patients with apathy have less favorable functional outcomes compared with their counterparts without apathy.

The recruitment details are shown in Figure 1. Participants will be recruited from patients admitted with a first-ever SAH to the neurosurgical centers of three regional hospitals: Prince of Wales Hospital, Kwong Wah Hospital, and Queen Mary Hospital. All patients with SAH (n = 320, approximately 50–60 per center per year) who are consecutively admitted to these centers over a 24-month period will be invited to participate. A research assistant (RA) will visit the neurosurgical wards weekly to identify eligible patients and obtain written consent. We anticipate that 75% of these patients will meet our inclusion criteria and agree to participate (52), resulting in approximately 240 potential participants [320 × (75%)]. Assuming a dropout rate of 13% (53), we expect that 209 [240 × (100% − 13%)] patients will complete the 15-month follow-up assessment.

There will be no patient or public involvement.

The proposed study is designed as a prospective cohort study. Written consent will be obtained from all patients or their next of kin. Demographic, psychosocial, and medical data will be collected from all patients. The number of and reasons for exclusions will be recorded. Details of the data collection schedule are provided in Table 1.

The neurological grade of patients at admission will be determined using the Glasgow Outcome Scale (54) and the WFNS grade scale. The WFNS scale categorizes patients from fully conscious with no motor or speech deficit (Grade I) to those with a Glasgow Coma Scale score of ≤6 (Grade V). Patients with a WFNS Grade I or II and Grade III or V will be classified as having good and poor grades, respectively.

For diagnostic assessments, all recruited patients will undergo delayed computed tomography (CT) scans of the brain 2 to 3 weeks after their initial presentation. Cerebral infarction due to delayed cerebral ischemia is defined as a new cerebral infarction on CT that cannot be attributed to any procedural causes (3). Clinical deterioration due to delayed cerebral ischemia is defined as presumably associated clinical deterioration that cannot be attributed to other potential causes (55). Hydrocephalus (acute or chronic) is characterized by persistent ventricular dilatation requiring the implantation of an internal shunt, such as a ventriculo-peritoneal shunt (3).

The severity of SAH on neuroimaging will be evaluated using the Fisher CT grade, which classifies findings into 4 levels: 1 for no blood visualized, 2 for diffuse or thin sheets (<1-mm thickness of vertical layers), 3 for localized clot and/or vertical layers (≥1-mm thickness), and 4 for the presence of intracerebral or intraventricular clots (56). In addition, details of the size of the ruptured aneurysm, its position (anterior cerebral artery, internal carotid/middle cerebral artery, and posterior circulation), the presence of intraventricular or intracerebral hemorrhage, and the treatment administered for the aneurysm will be recorded.

Apathy will be defined and quantified in accordance with the model proposed by Marin (57). An RA will operationally assess apathy using the Apathy Evaluation Scale (AES) (16, 58, 59), with an AES score of ≥37 indicating apathy (16). The study outcome will be the presence of apathy, categorized as either yes or no.

Three months after the onset of SAH (T1), patients and their caregivers will be assessed at a research clinic. This timing of the assessment is consistent with the schedule followed in other studies on apathy in patients with SAH (9) and stroke (16).

A trained RA will conduct interviews at a research clinic to assess apathy. The assessment will be based on interviews with the patients and their responses on the AES. This scale includes 18 items that evaluate the affective, behavioral, and cognitive domains of apathy by measuring motivational variables such as productivity, initiative, effort, emotional responsivity, novelty seeking or curiosity, perseverance, and social engagement. Responses are scored on a 4-point Likert-type scale, ranging from “not at all true/characteristic” to “very much true/characteristic,” with higher scores indicating more severe apathy. The AES has good reliability (58) and validity (60), and the internal consistency coefficient, test–retest reliability, and inter-rater reliability of the Chinese version of this scale are 0.90, 0.90, and 0.86, respectively (59). The AES was originally validated in a cohort with mixed diagnoses, namely, stroke, dementia, and depression (57), and has been widely used in various populations (60). To the best of our knowledge, no disease-specific measure of apathy has been developed for SAH or stroke. However, the AES has been used to assess apathy in patients with SAH (9, 24–26), and our group successfully used this scale to assess apathy in stroke (16).

A trained RA blinded to the patients’ apathy data will assess each patient’s level of functioning using the FIM (61), Lawton Instrumental Activities of Daily Living Scale (IADL) (62), Barthel Index (BI) (63), and modified Rankin Scale (mRS) (64). These measures have been previously used to measure functional outcomes after SAH (52, 65, 66). The FIM includes 18 items across 6 subscales: self-care, sphincters, mobility, communication, psychosocial, and cognition. Each item is rated on a 7-point scale from 1 (patient requires total assistance) to 7 (patient is completely independent). The overall score ranges from 18 to 126 points. The FIM is considered valid with good inter-rater agreement (67). The IADL scale is used to assess independent living skills, such as using a telephone, shopping, preparing food, performing housekeeping and laundry, transportation, managing medication responsibly, and handling finances. An IADL score of 18 is considered an excellent outcome. This scale has good content validity and excellent inter-rater reliability (0.99), test–retest reliability (0.90), and internal consistency (0.86) (62). The BI is used to measure independence based on 10 tasks, scored according to the amount of time or assistance required by the patient. The total BI score ranges from 0 to 20, with lower scores indicating greater dependency. The BI has favorable construct validity and predictive validity and moderate inter-and intra-reliability (0.4–0.6) (68). The mRS was previously used to rank outcomes (e.g., death, disability, and dependence) after stroke or SAH (53). The total mRS score ranges from 0 (no symptoms) to 6 (death), with 0 indicating an excellent outcome. In addition, a caregiver will complete a checklist on the patient’s post-SAH changes to assist with scoring, as recommended in a previous study (69). The construct validity and inter-rater reliability (0.6) of this checklist are excellent and moderate, respectively (68).

A trained RA will evaluate the levels of comorbidity, global cognitive functioning, and depressive symptoms using the Charlson Comorbidity Index (CCI) (70), Chinese version of the Beck Depression Inventory (BDI) (71), and Hong Kong version of the Montreal Cognitive Assessment (MoCA) (72), respectively. The CCI is the most widely used comorbidity adjustment method. It involves weighting 19 diseases from 1 to 6 points, and the sum of the weights is adjusted to produce final scores ranging from 0 to 33. This index has good validity and test–retest reliability and moderate-to-good inter-rater reliability (73). The BDI, consisting of 21 items, is scored on a 4-point Likert scale, with total possible scores ranging from 0 to 63. Higher scores indicate more severe depressive symptoms; specifically, scores of 0–9, 10–18, 19–30 and > 30 indicate no/minimal, minor/moderate, moderate/severe, and severe symptoms, respectively. The BDI has good internal consistency (0.86) (71), retest reliability (0.73–0.96), and validity (74). The MoCA is a one-page test with a maximum score of 30. It is used to evaluate seven cognitive domains, namely, visuospatial/executive functions, naming, verbal memory registration and learning, attention, abstraction, 5-min delayed verbal memory, and orientation. This scale has excellent discriminant validity, internal consistency (0.77), test–retest reliability (0.87), and inter-rater reliability (0.99) (72, 75). Any physiotherapy or cognitive/neuropsychological rehabilitation received by the patients will also be recorded.

All patients will undergo follow-up assessments of apathy and functioning at 9 (T2) and 15 months (T3) post-SAH or at 6 and 12 months after the first assessment. The AES, FIM, IADL, BI, mRS, MoCA, and BDI will be re-administered during the follow-up assessments.

Two hundred and forty patients will be recruited. Based on an estimated apathy frequency of 42% (7, 26), we expect to identify 101 (240 × 42%) cases of apathy. A sample size of 101 and 139 (i.e., 240–101) patients with and without apathy, respectively, would provide a power of 80% for the detection of any differences in functional outcomes between these 2 groups, assuming an effect size of apathy of ≥0.4. In a previous study, the effect size of apathy on functional outcomes in stroke survivors (Cohen’s d) was 2.4 (19).

Assuming a dropout rate of 13% (53), we anticipate that 88 (101 × 87%) patients with apathy will complete the follow-up visits. Although the rate of persistent apathy after SAH is unknown, previous studies on general stroke patients have reported persistent apathy rates ranging from 42 to 51% (27). Accordingly, our sample size is expected to achieve a power of 80% for the identification of predictors of persistent apathy with an odds ratio of ≥2.0. This is below the odds ratio of 3.3 that we previously reported for the impact of depressive symptoms on the 1-year course of apathy in stroke survivors (27). We will assume that the R² value of the other variables in the multivariate logistic regression (76) will be 0.21 (51).

All variables will be tested for normality using the Kolmogorov–Smirnov test, with a significance threshold of p < 0.05. The number and percentage of patients with persistent apathy will be calculated. To identify the predictors of persistent apathy, the baseline MoCA, BDI, and CCI scores as well as other demographic, clinical, and CT scan characteristics and physiotherapy or cognitive/neuropsychological rehabilitation received will be compared between patients with and without apathy at T2 using the χ2 test, Student’s t test, or Mann–Whitney U test as appropriate. A stepwise logistic regression will then be performed to assess the importance of the baseline MoCA, BDI and CCI scores, together with other significant variables identified in the univariate analyses. The same analyses will be repeated using data obtained at T3. We will also test a series of generalized estimating equation models to evaluate the associations of the baseline demographic, clinical, and CT scan characteristics with the risk of apathy across all follow-up assessments (T1/T2/T3). First, we will generate a univariate model to fit a logistic regression. Next, we will examine associations between demographic variables and the risk of apathy. The second model will include CT scan characteristics, whereas the final model will include the baseline MoCA, BDI, and CCI scores. The level of significance will be set at 0.05. We will perform two sub-analyses. First, the analysis will be repeated by substituting the MoCA total score with the executive functioning index score of the MoCA (77). Second, the analysis will be repeated for three subdomains (i.e., cognitive, affective, and behavioral) of the AES (78).

To examine the impact of apathy on functional outcomes, the FIM, IADL, BI, and mRS scores of the groups with and without apathy at T1 will be compared using a Student’s t test. The aforementioned outcome scores will then be adjusted for other putative outcome predictors (age, WFNS grade upon admission, intraventricular hemorrhage, and delayed cerebral ischemia) using analysis of covariance. Correlation of the functional outcome scores with the presence of apathy and other putative predictors will be computed using Pearson’s or Spearman’s correlation coefficients, as appropriate. Finally, the presence of apathy and significant predictors identified in the univariate analysis will be entered into a regression analysis of the functional outcome scores. This analysis will be repeated for T2 and T3.