Patient Participation in Medical and Social Decisions in Alzheimer's Disease

Johannes Hamann, MD; Katharina Bronner; Julia Margull; Rosmarie Mendel, PhD; Janine Diehl-Schmid, MD; Markus Bühner, PhD; Reinhold Klein, MD; Antonius Schneider, MD; Alexander Kurz, MD; Robert Perneczky, MD


J Am Geriatr Soc. 2011;59(11):2045-2052. 

In This Article


This cross-sectional study addressed the participation preferences and decisional capacity of patients with aMCI or mild dementia in AD. The survey consisted of interviews with patients, their relatives, and their referring physicians. Clinical data obtained during the diagnostic examination in the memory clinic where patients were recruited were also available.

Participating Patients

Individuals with aMCI or mild dementia in AD (Clinical Dementia Rating (CDR) 0.5 or 1[8]) were recruited from the outpatient memory clinic at the Department of Psychiatry and Psychotherapy, Technische Universität München, Munich, Germany. All patients had to fulfill National Institute of Neurological and Communicative Disorders and Stroke—Alzheimer's Disease and Related Disorders Association (NINCDS/ADRDA) criteria for probable AD[9] or revised consensus criteria for aMCI in conjunction with a relative's report on cognitive decline,[10] had to speak German fluently, and had to have a caregiver or relative consenting to be interviewed. Patients with evidence of other neurological or psychiatric causes of cognitive impairment, including frontotemporal lobar degeneration, Lewy-body disease, Parkinson's disease, alcohol dependence, traumatic brain injury, and depression, were excluded. The clinical diagnoses were established by consensus of two experienced clinicians. The routine diagnostic examination for all patients included interviews with the patients and the relatives, neuropsychological testing, physical and neurological examination, routine blood analysis, and structural brain imaging (magnetic resonance imaging or computed tomography). The psychometric assessment was based on the Consortium to Establish a Registry for Alzheimer's Disease (CERAD) test battery,[11] which includes the Mini-Mental-State Examination (MMSE).[12] All patients gave written informed consent after the study procedures had been explained in detail.

Participating Relatives

The relatives who had accompanied the patients to the diagnostic examination were also recruited for the study and also gave written informed consent after the study procedures had been explained in detail.

Participating Physicians (Referring Physicians)

The physicians who had referred patients to the memory clinic were surveyed in a telephone interview after patients had given permission to contact them.

Study Procedure and Review Board Approval

After written informed consent was obtained, one of two investigators (JM, KB) contacted patients and their relatives individually for the study. During the survey, the investigators went through the questionnaires task by task and assisted patients if they had difficulties understanding any of the items. Investigators were also instructed to pause if patients showed any sign of tiredness or lack of motivation.

The institutional review board of the medical faculty at Technische Universität München approved the study.

Data Obtained

Patients' age, sex, and years of education were recorded. Earlier studies have repeatedly found younger age, female sex, and higher education to be associated with preferences for greater participation.[13] Data on the psychometric assessment of the patients, the duration of their disease, and their current medications were obtained from the records.


Patients were surveyed about their participation preferences and their confidence in their decision-making capabilities. Finally, their decisional capacity was estimated using a semistructured interview—the MacArthur Competence Assessment Tool–Treatment (MacCAT-T[14]) for a hypothetical decision scenario.

The decision-making subscale of the Autonomy Preference Index (API), a six-item scale addressing patients' general wish to participate in medical decisions was used to measure patient participation preferences.[15]

For the API scale, totals were transformed to a range from 0 to 100, with 0 corresponding to complete lack of desire for autonomy (complete lack of desire to make decisions in the medical encounter) and 100 corresponding to the strongest possible desire to be an autonomous decision-maker in the medical context. For example, if a patient recorded the weakest preference for autonomy on all six questions, he or she would score 6 points, which corresponded to 0 on the adjusted scale. If he or she showed the strongest preference on each question, he or she would score 30 points, which corresponded to 100 on the adjusted scale.

Intermediate scores have been interpreted as reflecting a desire for decision-making shared equally between doctor and patient.[16]

A German version of the API was used, so independent persons translated the scale into German, retranslated it back into English, and sent to the authors of the original scales to avoid misunderstandings.[17]

Because the API refers to a decision-making dyad only (doctor, patient) and because previous research suggests that relatives (a third party) play an important role for patients with dementia, patients were asked to sort three cards ("patient," "relative," "doctor") according to who should have the greatest say concerning healthcare decisions. For the analysis, the mean ranks of the three parties ("patient," "relative," "doctor") were used. This approach was applied for general medical decisions and for four specific decisions (antihypertensive treatment, nursing home, car driving, and antidementia medication).

Patients were additionally asked to state their confidence in their decision-making ability (When you have medical problems, how confident are you concerning your ability to make good healthcare decisions?) on a 5-point scale ranging from not confident at all to completely confident.

Patients' insight into their illness was captured using the Clinical Insight Rating (CIR) Scale,[18] which assesses awareness of four facts of insight regarding dementia (reason for the visit to doctor, cognitive deficits, functional deficits, perception of progression of the disease). Each item is ranked from 0 to 2, yielding a total rating from 0 (insight fully preserved) to 8 (insight totally absent). Research assistants assigned ratings on this scale based on their judgments of patients' insight after the interviews with the patient and the relative.

The MacCAT-T was used to study patients' decisional capacity.[14] The MacCAT-T is designed as a structured approach to assesses individuals' competence to make treatment decisions by examining their capacities in four areas (understanding information, reasoning about the potential risks and benefits of a treatment, appreciating the nature of their situation and the consequences of a treatment choice, and expressing a choice). Within the MacCAT interview, patients were provided with information about a hypothetical therapeutic decision (treatment with a vaccination against AD, see online Appendix S1). The course of the treatment (e.g., injections) and advantages and disadvantages of the treatment were explained to patients. Understanding was then assessed by exploring the patients' ability to paraphrase what had been disclosed to them concerning the course of the recommended treatment and the treatment's pros and cons.

Appreciation was assessed by exploring whether the patient acknowledged that the proposed treatment might have at least some benefit (the appreciation of treatment benefit subscale).

Reasoning was assessed by asking patients to explain their choices (whether they mentioned any consequences of undergoing the treatment (or not), whether they compared the alternatives (treatment vs no treatment), whether they expressed any thoughts about consequences other than those offered in the disclosure, and whether their final choice followed logically from their own explanations). Finally, the ability to express a choice was assessed by recording whether the patient stated a preference for a treatment option.[19] The original MacCAT-T was adapted to the specifics of the current study. Thus, because patients had different diagnoses (AD and MCI) with different prognoses, the understanding disorder and appreciation of disorder sections were not included. As previously suggested,[14] sum scores for the understanding sections were constructed by adding up single item scores and dividing by the number of items. This resulted in two subscores ranging from 0 to 2 (understanding treatment, understanding risks and benefits). The sum score for understanding then ranges from 0 to 4.

Two researchers (KB, JM) had been trained with a videotape before the start of the study to perform the MacCAT interview in German.

Within the decision scenario, patients were again asked for their participation preferences (card sorting) and their confidence in their decision-making capability (5-point scale).

Patients' relatives were surveyed regarding their estimate of the patients' participation preferences (card sorting) and decisional capacity (5-point scale). Likewise, patients' referring physicians were surveyed in a telephone interview regarding their estimate of the patients' participation preferences and their decisional capacity. Relatives and physicians were surveyed about their estimates concerning general health decisions and the hypothetical decision vignette (vaccination against AD) used for the MacCAT-T interview.

Analysis of the Data

Patients with aMCI and mild dementia in AD were pooled for the analyses to increase the variance in the data set. aMCI is often a predementia AD stage, with yearly progression rates to dementia in AD of up to 20%.[20] Descriptive statistics (means and frequencies) were used to characterize the sample. Patient participation preferences and all other ratings were described using means and standard deviations. An ordinal regression analysis with patient participation preferences (rank) as dependent variable and age, sex, education (≤10 vs >10 years), MMSE score, and patients' confidence in their decision-making abilities as independent variables was used to predict patient participation preferences. The Wilcoxon rank test was used to compare the relatives' and doctors' ratings with the patients' actual participation preferences, and Kendall's tau (b) was calculated to assess whether relatives and doctors could predict patients' participation preferences. T-tests were used to compare relatives' and physicians' ratings of patients' decisional capacity with patient self report.

Pearson correlations were used to correlate the different measures for decisional capacity.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.