¶ Membership of the PRODEM Study Group is provided in the Acknowledgments.
The authors have declared that no competing interests exist.
Conceived and designed the experiments: RS HS TB GR PDB CE MU JM FL PK CB. Performed the experiments: RS TB GR PDB CE MU JM FL PK CB AL SS PS GS RL BH. Analyzed the data: SS HS RS PL. Wrote the paper: SS RS HS PL GR TB.
To assess the influence of cognitive, functional and behavioral factors, co-morbidities as well as caregiver characteristics on driving cessation in dementia patients.
The study cohort consists of those 240 dementia cases of the ongoing prospective registry on dementia in Austria (PRODEM) who were former or current car-drivers (mean age 74.2 (±8.8) years, 39.6% females, 80.8% Alzheimer’s disease). Reasons for driving cessation were assessed with the patients’ caregivers. Standardized questionnaires were used to evaluate patient- and caregiver characteristics. Cognitive functioning was determined by Mini-Mental State Examination (MMSE), the CERAD neuropsychological test battery and Clinical Dementia Rating (CDR), activities of daily living (ADL) by the Disability Assessment for Dementia, behavior by the Neuropsychiatric Inventory (NPI) and caregiver burden by the Zarit burden scale.
Among subjects who had ceased driving, 136 (93.8%) did so because of “Unacceptable risk” according to caregiver’s judgment. Car accidents and revocation of the driving license were responsible in 8 (5.5%) and 1(0.7%) participant, respectively. Female gender (OR 5.057; 95%CI 1.803–14.180; p = 0.002), constructional abilities (OR 0.611; 95%CI 0.445–0.839; p = 0.002) and impairment in Activities of Daily Living (OR 0.941; 95%CI 0.911–0.973; p<0.001) were the only significant and independent associates of driving cessation. In multivariate analysis none of the currently proposed screening tools for assessment of fitness to drive in elderly subjects including the MMSE and CDR were significantly associated with driving cessation.
The risk-estimate of caregivers, but not car accidents or revocation of the driving license determines if dementia patients cease driving. Female gender and increasing impairment in constructional abilities and ADL raise the probability for driving cessation. If any of these factors also relates to undesired traffic situations needs to be determined before recommendations for their inclusion into practice parameters for the assessment of driving abilities in the elderly can be derived from our data.
With few exceptions
Practice parameters for medical assessment are needed, but they are scarce and at this point rather rely on expert opinion than evidence-based parameter selection
The Canadian consensus conference on dementia suggests cognition, function and medical status being important in the evaluation of driving abilities
We also assessed the influence of Behavioral and Psychological Symptoms of Dementia (BPSD) as well as caregiver characteristics and strain of care on driving cessation.
The study was approved by the ethics committees of the Medical University of Graz, the Medical University of Innsbruck, the Medical University of Vienna, the Konventhospital Barmherzige Brüder Linz, the Province of Upper Austria, the Province of Lower Austria and the Province of Carinthia. Written informed consent was obtained from all patients and their caregivers.
The prospective registry on dementia in Austria (PRODEM) is an ongoing longitudinal multi-center cohort study conducted in 12 memory clinics in our country. Since 2009, 437 subjects have been included. Inclusion criteria are (1) dementia diagnosis according to DSM-IV criteria
The study centers were situated in six of nine provinces of the State of Austria with investigators representing the specialties of neurology and/or psychiatry. Historical information and clinical as well as neuropsychological examinations were collected at baseline and every six months over a time period of two years or until institutionalization, withdrawal from the study, loss to follow-up or death. At each visit, patient- and caregiver assessments followed a pre-defined protocol, which was administered at every participating center. The baseline evaluation included patient- and caregiver demographics, duration of dementia symptoms, assessment of the patients’ living situation and resource utilization, driving status, presence of co-morbidities, recording of anti-dementia and concomitant medication, as well as extensive clinical, cognitive, behavioral and functional assessment. Caregiver burden was also assessed. Bio-banking including sampling of DNA, RNA, plasma and serum was done. MRI scans were obtained according to standardized protocols.
The current study cohort consists of those 240 study participants who ever drove a car. Alzheimer’s disease (AD), Vascular Dementia (VaD), Lewy Body Dementia (LBD), Frontotemporal Dementia (FTD) and other dementias were diagnosed in 194, 12, 11, 16 and 7 patients, respectively. Their mean age was 74.2 (±8.8) years and ranged from 41 to 100 years. There were 95 (39.6%) females. The mean duration of dementia prior to inclusion into the study was 33.0 (±24.6) months. The mean MMSE and CDR scores of patients were 22.1 (±4.5) and 0.9 (±0.6), respectively. The mean DAD score was 71.5% (±25.3%). At study entry 145 (60.4%) participants had ceased driving during the course of dementia.
In patients with possible AD, we used MRI findings to determine evidence for mixed AD and vascular pathology. From a total of 61 patients with possible AD, MRI was available in 41 subjects. Among them, 9 (22%) patients had evidence for mixed pathology.
Causes for driving cessation were obtained on the basis of information from the caregivers. They were categorized into (1) unacceptable risk, (2) involvement in a car accident, or (3) revocation of the driver’s license.
Dementia types were diagnosed according to the NINCDS-ADRDA Criteria for AD
Age, sex, retirement- and marital status, education, occupation and living situation including assistance at home were assessed by structured questionnaires at study entry. Educational level was categorized into (1) less than high school diploma, (2) high-school diploma and (3) university degree. Occupational status was classified according to the patients’ longest occupation in life. Categories were (1) blue-collar worker, (2) white-collar worker, (3) self-employment and (4) housewife.
We assessed the Mini Mental State Examination (MMSE)
For assessment of cognitive functioning the “Consortium to Establish a Registry for Alzheimer’s disease (CERAD) – Plus” neuropsychological test battery has been used
To assess function, we applied the disability assessment for dementia (DAD) scale
Assessment of medical status included history of stroke, coronary heart disease, atrial fibrillation, venous thrombosis and major vascular risk factors including hypertension, diabetes and hypercholesterolemia. Definition of vascular risk factors followed the American College of Cardiology Foundation/American Heart Association
BPSD were assessed by the Neuropsychiatric Inventory (NPI)
The relationship between caregiver and patient was evaluated and categorized into spouse, unmarried partner, child, other relative or non-relative. The caregivers’ employment status distinguished between unemployed, fully employed, partly employed, in training, reduced employment because of care, retired or other.
Caregiver burden was assessed using the Zarit Burden Interview (ZBI)
For statistical analysis we used the Statistical Package of Social Sciences (SPSS) version 19. Between-group differences in categorical variables were compared using the χ2 test. Assumptions of normal distribution for continuous variables were tested with the Kolmogorov-Smirnov test. Normally distributed continuous variables were compared using the Student’s t-test and the Mann-Whitney-U test was applied in case of non- normally distributed variables. All patient- and caregiver related factors found to be associated with driving cessation at a p-value lower than 0.05 in univariate statistical analysis were then simultaneously entered into a multivariate logistic regression model to determine their independent contribution on the patients’ driving status. Variables were assessed for multicollinearity by Pearson’s and Spearman’s correlation coefficient and evaluated for exclusion if values >0.7 occurred. Odds ratios and 95% Confidence Intervals were calculated from the beta coefficients and their standard errors.
With 90.9% the highest rate of driving cessation was seen in patients with LBD. The cessation rate in AD cases was 58.2%. Similar figures were seen in VaD and FTD with 66.7% and 56.3%, respectively. The subgroup of study participants with evidence for a mixed AD and vascular pathology ceased driving at a comparable rate of 55.6%.
In 136 cases (93.8%) the cause for driving cessation was “too high risk” reported by the caregiver. Car accidents caused cessation in 8 cases (5.5%) and revocation of the driving license in only 1 participant (0.7%). As can be seen from
Variable | Driving cessation | No driving cessation | p-value |
N = 145 | N = 95 | ||
Age, yrs (median, IQR) | 77(71.00–81.00) | 74(67.00–79.00) | 0.037 |
Female gender (N, %) | 65(44.80) | 30(31.60) | 0.040 |
Retired (N, %) | 132(97.10) | 80(90.90) | 0.046 |
Assistance at home (N, %) | 126(87.50) | 73(76.80) | 0.031 |
MMSE (median, IQR) | 22.00(18.50–25.00) | 24.00(21.00–26.00) | <0.001 |
CDR (median, IQR) | 1.00(0.50–1.00) | 0.50(0.50–1.00) | <0.001 |
CERAD Constructional practice (mean, SD) | −0.98 (1.79) | −0.13(1.51) | 0.001# |
NPI – Aberrant motor behavior (N, %) | 34(23.40) | 12(12.80) | 0.041 |
NPI – Apathy (N, %) | 66(45.80) | 28(29.80) | 0.013 |
DAD (median, IQR) | 67.50(42,50–87.50) | 87.00(72.50–97.50) | <0.001 |
ZBI (median, IQR) | 22(11.00–32.00) | 13(4.00–24.00) | <0.001 |
SD = Standard Deviation, IQR = interquartile range.
Mann-Whitney-U-Test.
χ2 Test. #Student’s t-test. Abbreviations: MMSE = Mini Mental State Examination, CDR = Clinical dementia rating, CERAD = Consortium to establish a registry for Alzheimer’s disease, NPI = Neuropsychiatric Inventory, DAD = Disability Assessment for Dementia, ZBI = Zarit Burden Interview.
Logistic regression analysis determined female gender, constructional abilities on CERAD and impairment in ADL to be the only significant and independent associates of driving cessation (
Variable | OR | 95% CI | P-value |
Age | 1.048 | 0.986–1.114 | 0.129 |
Female gender | 5.057 | 1.803–14.180 | 0.002 |
Retired | 1.627 | 0.257–10.285 | 0.605 |
Assistance at home | 0.721 | 0.234–2.225 | 0.569 |
MMSE | 1.034 | 0.896–1.194 | 0.646 |
CDR | 1.553 | 0.343–7.042 | 0.568 |
CERAD constructional practice | 0.611 | 0.445–0.839 | 0.002 |
NPI – aberrant motor behavior | 1.524 | 0.417–5.571 | 0.524 |
NPI – apathy | 0.521 | 0.191–1.424 | 0.204 |
DAD | 0.941 | 0.911–0.973 | <0.001 |
ZBI | 1.027 | 0.989–1.066 | 0.164 |
Abbreviations: OR = Odds Ratio, CI = Confidence Interval, MMSE = Mini Mental State Examination, CDR = Clinical dementia rating, CERAD = Consortium to establish a registry for Alzheimer’s disease, NPI = Neuropsychiatric Inventory, DAD = Disability Assessment for Dementia, ZBI = Zarit Burden Interview.
We confirm previous studies reporting that one in three dementia patients still drives
Others also reported that women are more likely to cease driving than men. This was observed in healthy aging
Unlike other studies
There was also no association between driving status and dementia type, but the relatively low number of non-Alzheimer dementias in our investigation needs to be emphasized. Our study is the first to include caregiver related factors as possible predictors of driving cessation in demented patients. The study result that the caregivers’ estimate of “unacceptable risk” was the reason to cease driving in more than 90 percent of our patients underscores the pivotal role and key responsibility of caregivers in the decision as to whether demented patients still drive. The reliability of the caregivers’ judgment regarding fitness to drive in old people is supported by a recent study by Stapleton and coworkers
A strength of our study is its prospective study design with the use of pre-defined and standardized questionnaires in a nation-wide multicenter setting.
There are also weaknesses. We included only patients who attended memory clinics and who had caregivers willing to be part of the investigation. Therefore the cohort composition may not be representative for the general dementia population. It is likely that we rather overestimated the frequency of driving cessation given the central role of caregivers in the decision-making as to whether demented patients quit driving.
We can also not exclude with certainty that the factors that influence driving cessation in dementia patients with caregivers differ from those in patients without caregivers. It is also possible that other somatic factors that were not specifically included in our analysis like motor dysfunction or visual disturbance influenced driving status. Other factors which remained un-assessed in our investigation despite they have been shown to be related to performance on structured road tests are measures of strategic and tactical thinking
We consider the identification of demographic, clinical and caregiver-related associates of driving cessation in patients with dementia only the first step in the development of evidence-based practice guidelines. Future studies determining the role of each single factor or their combination as predictors of moving violations including car accidents are warranted.
PRODEM Study Group co-investigators:
Medical University of Graz: Reinhold Schmidt, Erich Flooh, Paul Freudenberger, Anja Grazer, Anita Harb, Edith Hofer, Elfi Hofer, André Fixa, Christine Gadhery, Nina Homayoon, Anita Lechner, Patricia Linortner, Marisa Loitfelder, Cristoph Murgg, Katja Petrovic, Irmgard Poelzl, Birgit Reinhart, Michael Schallert, Helena Schmidt, Semmler-Bruckner Josef, Spechtl Michael, Stephan Seiler, Abhijit Sen, Anna Toeglhofer
Medical University of Innsbruck: Thomas Benke, Margarete Delazer, Guenter Sanin
General Hospital Linz: Gerhard Ransmayr, Christoph Arzt, Alexandra Fuchs, Michael Guger, Christine Hoflehner, Andrija Javor, Riccarda Lehner, Brigitta Neubauer, Susanne Schmidegg, Michaela Steffelbauer, Walter Struhal
Medical University of Vienna: Peter Dal-Bianco, Michaela Imre, Agnes Pirker, Peter Santer, Evelyn Sieczkowski,
Konventhospital der Barmherzigen Brüder Linz: Christian Eggers, Joachim Adl, Bernhard Haider
Landesnervenklinik Sigmund Freud Graz: Margarete Uranues, Jasmin Weber
Regional Hospital Hall in Tirol: Josef Marksteiner, Angela Diwo, Douglas Imarhiagbe
Nervenklinik Wagner-Jauregg Linz: Friedrich Leblhuber, Beran-Praher Margit, Szalay Elisabeth
Regional Hospital Villach: Peter Kapeller, Ingolf Koechl
Regional Hospital Horn: Christian Bancher, Gustav Feldner, Takeshi Nakajima