Study and participants
The Swedish Alzheimer Treatment Study (SATS) is a 3-year, prospective, open, non-randomized multicentre study with the purpose of assessing long-term ChEI treatment (donepezil, rivastigmine and galantamine) in routine clinical practice. Several publications have previously reported various findings from the SATS [2, 6, 7, 14]. In total, 1258 participants with AD were recruited from 14 memory clinics in different areas of Sweden. Of these, 734 individuals were defined as having mild AD (Mini-Mental State Examination (MMSE)  score, 20–26) and 287 individuals as having moderate AD (MMSE score, 10–19) at the start of ChEI therapy (baseline) and were included in the present study.
The inclusion criteria were outpatients aged 40 years and older who received a clinical diagnosis of dementia as defined by the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (DSM-IV)  and possible or probable AD according to the criteria of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA) . In addition, the participants were required to be community dwelling with or without home-help services at the time of AD diagnosis, to have a responsible caregiver and to be capable of being evaluated using the MMSE scale at the initiation of ChEI treatment. Exclusion criteria were not fulfilling the diagnostic criteria for AD, already receiving active ChEI therapy or contraindications to ChEI. Concomitant medications other than ChEIs were recorded at baseline and allowed during the study, except for memantine. If memantine therapy was commenced, the individual dropped out from the SATS at that time point.
The SATS participants were investigated in a structured, follow-up programme over 3 years that investigated cognition, global performance and instrumental and basic ADL abilities, at the start of ChEI treatment, after 2 months (MMSE and global rating only) and every 6 months. Nurses trained to care for patients with dementia assessed the ADL capacity based on interviews with the caregiver. The dates of eventual nursing home placement and death were documented, as well as the date of, and reason for, any withdrawal from the SATS.
After inclusion in the study and the baseline evaluations, the participants were prescribed ChEI therapy as part of the ordinary Swedish health care system, in accordance with the approved product labelling. The SATS is an observational study and the choice of ChEI agent and dose was left entirely up to the dementia specialist’s discretion and professional judgement. The ChEI dose was recorded after 2 months of treatment and then semi-annually after baseline. If the patient stopped taking the ChEI, the individual was excluded from the study at that time point.
Ethics, consent and permissions
All patients and/or their caregivers gave their written informed consent to participate in the SATS, which was conducted according to the provisions of the Helsinki Declaration and was approved by the Ethics Committee of Lund University, Sweden.
Cognitive status was assessed using the MMSE scale (with scores ranging from 0 to 30; a lower score indicates more impaired cognition) and the Alzheimer’s Disease Assessment Scale—cognitive subscale (ADAS-cog)  (0–70 points; a lower score indicates higher cognitive ability). The Clinician Interview-Based Impression of Change (CIBIC)  was used as a global rating of ‘change from the start of ChEI treatment’. The evaluations were performed at all intervals using a 7-point scale that varied from 1 (very much improved) to 7 (marked worsening). Three groups of response were defined at each CIBIC interval: 1–3 indicated improvement, 4 indicated no change and 5–7 indicated worsening. No guidelines or descriptors were provided to define the individual ratings. The classification between, for example, minimally improved or very much improved was left to the dementia specialist’s clinical judgement.
The functional capacity was assessed using the Instrumental Activities of Daily Living (IADL) scale , which comprises eight items: ability to use the telephone, shopping, food preparation, housekeeping, ability to do laundry, mode of transportation, responsibility for own medications and ability to handle finances. Each item was scored from 1 (no impairment) to 3–5 (severe impairment), thus allowing a total range of 8–31 points. Basic ADL was measured by the Physical Self-Maintenance Scale (PSMS)  comprising six items: toilet, feeding, dressing, grooming, physical ambulation and bathing. Each item was scored from 1 (no impairment) to 5 (severe impairment), thus allowing a total range of 6–30 points.
For each follow-up visit, we calculated the mean MMSE, ADAS-cog, IADL and PSMS changes from baseline with 95 % confidence intervals (CI). To facilitate comparisons of these rates, we converted the change in score to positive values (indicating improvement) and negative values (indicating worsening). The proportions of improved/unchanged SATS patients, predefined as those who demonstrated an improvement or no change (≥0 points difference) at the respective evaluation, were also calculated for the MMSE, ADAS-cog, IADL and PSMS scales.
Nursing home placement was defined as the permanent admission to a licensed skilled nursing facility with 24-hour care; that is, rehabilitative or respite care was not included. If hospitalization occurred before nursing home entry, the date of hospital admission was used. Using the 12-digit personal identity number assigned to each resident of Sweden, we determined whether each participant in the study was still alive on 31 December 2013 with the help of the Swedish population register (Swedish Tax Agency). If not, the date of death was recorded.
The IBM Statistical Package for the Social Sciences (SPSS) for Windows (version 22.0; IBM Corporation, Armonk, NY, USA) was used to perform the statistical analyses. The level of significance was defined as p <0.05, unless otherwise specified, and all tests were two-tailed. Observed-case analyses were used to avoid overestimation of the therapeutic effect by imputing earlier, better outcome scores in a long-term study of a progressively deteriorating disease. Parametric tests were used because of the large sample size and the approximately normally distributed continuous potential predictors. Independent-sample t tests were used to compare the differences between the means for two groups, and chi-square tests were conducted to analyse categorical variables. Pearson’s correlation coefficient was calculated to investigate any linear associations between continuous variables.
Mixed, linear and non-linear fixed and random coefficient regression models using the subject as a hierarchical variable (to consider the intra-individual correlation) were performed. In addition, the mixed-effects models took into account the varying number of assessments available for each patient and unequal time intervals between follow-ups, which are the usual concerns in longitudinal studies. The individuals who discontinued the study contributed information during their time of participation; hence, we considered the trajectories of all patients in the SATS.
Time was defined as the exact number of months between the start of ChEI therapy and each visit, which implies that all data points were used at the actual time intervals. To adjust for baseline differences, the initial cognitive, instrumental or basic ADL scores for each individual and their interaction with linear and quadratic terms for time in the study (to enable a non-linear rate of change in the models) were included as fixed effects; that is, time in months (and time in months2) × MMSE (ADAS-cog, IADL or PSMS) baseline score. Thus, the dependent variables were the cognitive or functional scores assigned at the second and subsequent evaluations for each participant; the mixed-effects models do not intend to predict the scores at the initiation of ChEI treatment. The random terms were an intercept and time in months, with a variance components covariance matrix. Several potential socio-demographic and clinical predictors were included as fixed effects in the models, such as sex, age at the start of ChEI therapy, clinician’s estimate of age at AD onset, years of education, presence of the apolipoprotein E (APOE) ε4 allele (no/yes), solitary living (no/yes), number of medications at baseline, and specific concomitant medications (no/yes for each group) including antihypertensive/cardiac therapy, antidiabetics, asthma medication, thyroid therapy, lipid-lowering agents, oestrogens, non-steroidal anti-inflammatory drugs (NSAIDs)/acetylsalicylic acid, antidepressants, antipsychotics and anxiolytics/sedatives/hypnotics. The effect of ChEIs was analysed using the drug agents (coded as a set of dummy variables) and dosages. The ChEI dose could vary during the treatment period for an individual patient and between patients; therefore, the mean dose used during the entire follow-up period was calculated for each participant. In cases of drop-out, the mean dose used during the individual’s time of participation in the SATS was calculated. To obtain a similar metric for percentage maximum dosage for each of the three ChEIs, the mean dose was divided by the maximum recommended dose for each drug; that is, 10 mg for donepezil, 12 mg for rivastigmine (oral administration) and 24 mg for galantamine. The term ‘ChEI agent × dose’ was also included in the models. Furthermore, some potential interactions (gender, age or education) with disease severity at baseline or with time in the study were included in the models. Non-significant variables (p >0.05) were eliminated in a backward stepwise manner. The hierarchical principle was applied in the mixed-effects models; variables that appeared in significant interactions were not considered for elimination.