Health care for older people in Italy: The U.L.I.S.S.E. Project (Un link informatico sui servizi sanitari esistenti per l'anziano - a computerized network on health care services for older people).

OBJECTIVES: The U.L.I.S.S.E. study is aimed at describing older patients who are cared for in hospitals, home care or nursing homes in Italy. DESIGN: The U.L.I.S.S.E. study is an observational multicenter prospective 1-year study. SETTING: Overall, 23 acute geriatric or internal medicine hospital units, 11 home care services and 31 nursing homes participated in the study. MEASUREMENTS: The patient's evaluation was performed using comprehensive geriatric assessment instruments, i.e. the interRAI Minimum Data Set, while data on service characteristics were recorded using ad-hoc designed questionnaires. RESULTS: The older subjects who are in need of acute and long term care in Italy have similar characteristics: their mean age is higher than 80 years, they have a high level of disability in ADL, an important multimorbidity, and are treated with several drugs. The prevalence of cognitive impairment is particularly high in nursing homes, where almost 70% of residents suffer from it and 40% have severe cognitive impairment. On the other hand, there is a shortage of health care services, which are heterogeneous and fragmented. CONCLUSIONS: Health care services for older people in Italy are currently inadequate to manage the complexity of the older patients. An important effort should be undertaken to create a more integrated health care system.

Association between ACE inhibitors use and headache caused by nitrates among hypertensive patients: results from the Italian group of pharmacoepidemiology in the elderly (GIFA).

Treatment with ACE inhibitors has shown to be effective in the prophylaxis of migraine attacks. The aim of this study was to explore whether among hospitalized hypertensive patients use of ACE inhibitors may reduce the risk of headache caused by nitrates. To this end, we used the GIFA database, that includes patients admitted to academic medical centres throughout Italy. We studied 1537 patients (mean age 75 +/- 10 years) receiving treatment with nitrates during a hospital stay and diagnosed with hypertension. Headaches that had a probable or definite causal relation with nitrates use based on the Naranjo algorithm were considered for this analysis. Of the total enrolled sample, 762 patients (50%) used ACE inhibitors during hospital stay. Headache caused by nitrates was recorded in 12/762 (1.6%) ACE inhibitor users and in 24/775 (3.2%) other participants (P = 0.049). After adjusting for potential confounders, ACE inhibitors use was associated with a significantly lower risk of headache (OR 0.43; 95% Confidence Intervals: 0.20-0.90). This result was confirmed if ACE inhibitors use was compared with use of other antihypertensive agents (OR 0.44; 95% CI 0.20-0.95). In conclusion, this study suggests that among hypertensive subjects use of ACE inhibitors is associated with a reduced risk of headache caused by nitrates.

Lifestyle-related risk factors for Parkinson's disease: a population-based study.

OBJECTIVES: To investigate the association of major lifestyle-related risk factors with the prevalent cases of Parkinson's disease (PD) identified by the Italian Longitudinal Study on Aging. METHODS: A total of 5632 individuals randomly selected from the population registers of eight centers were screened for parkinsonism using both a questionnaire and a neurologic examination. Screened positives underwent a structured clinical work-up for the diagnosis of parkinsonism and parkinsonism subtypes. RESULTS: We identified 113 prevalent cases of PD. Age, male gender, and pesticide-use license were significantly related to PD. Heavy smoking was inversely related to PD. Age (OR = 1.1; 95% CI, 1.06-1.15) and pesticide-use license (OR = 3.7; 95% CI, 1.6-8.6) kept their significant correlation with the disease in the multivariate analysis to adjust for all the variables under investigation. Multivariate analyses were made for men and women separately: pesticide exposure was positively associated with PD only in men. CONCLUSIONS: Pesticide exposure might represent a candidate for environmental factors involved in PD.

Construct validity of the 15-item geriatric depression scale in older medical inpatients.

The construct validity of the 15-item Geriatric Depression Scale (sfGDS) has been assessed in selected populations. The aim of this study was to assess the appropriateness of applying the sfGDS to unselected older inpatients. The main component analysis of sfGDS was performed in 2032 medical inpatients (mean age = 76.3 +/- 8.4). sfGDS did not qualify as a unidimensional test. Three factors explained 47.7% of variance and explored the following dimensions: positive attitude toward life, distressing thoughts/negative judgment about the own condition, and inactivity/reduced self-esteem. The internal homogeneity was poor (Cronbach's alpha = .46). A higher fraction of variance was explained in patients independent in all or dependent in > or = 1 activity of daily living (ADL). In older medical inpatients, sfGDS is not a single construct, which prevents the univocal interpretation of the final score. The higher fraction of explained variance in patients with comparable ADL performance probably reflects the dependency of affective from physical status.

Construct validity of the abbreviated mental test in older medical inpatients.

OBJECTIVES: To evaluate validity and internal structure of the Abbreviated Mental Test (AMT), and to assess the dependence of the internal structure upon the characteristics of the patients examined. DESIGN: Cross-sectional examination using data from the Italian Group of Pharmacoepidemiology in the Elderly (GIFA) database. SETTING: Twenty-four acute care wards of Geriatrics or General Medicine. PARTICIPANTS: Two thousand eight hundred and eight patients consecutively admitted over a 4-month period. MEASUREMENTS: Demographic characteristics, functional status, medical conditions and performance on AMT were collected at discharge. Sensitivity, specificity and predictive values of the AMT <7 versus a diagnosis of dementia made according to DSM-III-R criteria were computed. The internal structure of AMT was assessed by principal component analysis. The analysis was performed on the whole population and stratified for age (<65, 65-80 and >80 years), gender, education (<6 or >5 years) and presence of congestive heart failure (CHF). RESULTS: AMT achieved high sensitivity (81%), specificity (84%) and negative predictive value (99%), but a low positive predictive value of 25%. The principal component analysis isolated two components: the former component represents the orientation to time and space and explains 45% of AMT variance; the latter is linked to memory and attention and explains 13% of variance. Comparable results were obtained after stratification by age, gender or education. In patients with CHF, only 48.3% of the cumulative variance was explained; the factor accounting for most (34.6%) of the variance explained was mainly related to the three items assessing memory. CONCLUSION: AMT >6 rules out dementia very reliably, whereas AMT <7 requires a second level cognitive assessment to confirm dementia. AMT is bidimensional and maintains the same internal structure across classes defined by selected social and demographic characteristics, but not in CHF patients. It is likely that its internal structure depends on the type of patients. The use of a sum-score could conceal some part of the information provided by the AMT.

Exacerbated chronic obstructive pulmonary disease: a frequently unrecognized condition.

OBJECTIVES: To assess to which extent exacerbated chronic obstructive pulmonary disease (COPD) remains unrecognized in the emergency department, which factors account for misdiagnosis and which are the effects of misdiagnosis on patient's management. DESIGN: Retrospective study and stratified random sampling method as selection criterion. SETTING: University Hospital. SUBJECTS: Eighty patients representative of those discharged from the wards of medicine with a diagnosis of exacerbated COPD and 72 having a discharge diagnosis of coronary artery disease (CAD) were studied. MAIN OUTCOME MEASURES: Degree of concordance between admission and discharge diagnosis; presenting symptoms and signs of patients correctly or incorrectly classified on admission; impact of diagnostic procedures carried out by the physician on call on patient's management. RESULTS: The correct diagnosis was missed on admission in 13/80 COPD and 3/72 CAD patients (chi(2): 5.87, P=0.015). The prevalence of the following presenting features distinguished the 67 COPD patients who were correctly classified on admission from the remaining 13: severe weakness (21 vs. 10, chi(2): 9.53, P=0.002), dyspnea (60 vs. 3, chi(2): 28.75, P < 0.001), and limb oedema (14 vs. 6, chi(2): 3.70, P=0.054). Critical hypoxemia was diagnosed and treated only after admission in 19 COPD patients. CONCLUSIONS: Exacerbated COPD frequently escapes recognition in the emergency room, mainly if severe weakness and limb oedema are its presenting features. Arterial blood gas analysis is not systematically performed in the emergency room and, consequently, oxygen therapy is either not administered or given to selected COPD patients on an empirical basis.

Trends in prescribing ACE-inhibitors for congestive heart failure in elderly people.

BACKGROUND AND AIMS: The aim of this study was to analyze trends in the use of ACE-inhibitors in patients aged 65 and older with congestive heart failure (CHF) in the period from 1988 to 1998. METHODS: We studied 2985 patients (mean age 79.7 +/- 7 years), hospitalized for CHF in 12 different bimonthly periods. Home therapy prior to hospitalization was assessed retrospectively, and data on in-hospital therapy and discharge prescriptions were collected prospectively. RESULTS: Diuretics and digitalis were the most commonly used and prescribed drugs. The use of ACE-inhibitors between 1988 and 1998 increased from 13.4 to 46.7% prior to hospitalization, and from 25.8 to 59.2% as a discharge prescription. The most important factors associated with a prescription of ACE-inhibitors at discharge were previous use (OR 4.35, 95% CI=3.65-5.19), hypertension (OR 1.76, 95% CI=1.47-2.11), valvular heart diseases (OR 2.06, 95% CI=1.51-2.81) and diabetes (OR 1.58, 95% CI=1.29-1.93). Physical impairment was associated with a decreased use of ACE-inhibitors at discharge (OR 0.55, 95% CI=0.45-0.67). CONCLUSIONS: The use of ACE-inhibitors for the treatment of CHF progressively increased both at home and in hospital wards of general medicine and geriatrics in the 10-year period studied. Nevertheless, digitalis and diuretics continue to be the most commonly prescribed drugs. A widespread educational effort is needed to increase physicians' awareness of the rationale for prescribing ACE-inhibitors for CHF patients.

Abdominal aortic aneurysm in normotensive patients: association with angiotensin-converting enzyme gene polymorphism.

OBJECTIVE AND DESIGN: to assess if deletion of the angiotensin-converting enzyme (ACE) gene is a risk factor for abdominal aortic aneurysms (AAAs) in normotensive patients. MATERIALS AND METHODS: ACE gene polymorphism was examined by polymerase chain reaction in 124 subjects with AAA and in 112 control subjects. AAA normotensive patients (group A, n=56) were compared to normotensive control subjects (group B, n =112) and to AAA hypertensive patients (group C, n =68). All subjects enrolled in this study were Caucasian and from central and southern Italy. RESULTS: the distribution of ACE genotypes was: normotensive patients with AAAs (group A): 3 II, 14 ID, 39 DD; normotensive control subjects (group B): 36 II, 48 ID, 28 DD; hypertensive patients with AAAs (group C): 14 II, 32 ID, 22 DD. The DD genotype was more common in group A than in control groups (A vs B p<0.001; A vs C p <0.001). The ID genotype was more common in group A as well (A vs B p <0.05; A vs C p <0.005). CONCLUSIONS: our data suggest a role for ACE I/D gene polymorphism in the pathogenesis of AAA in normotensive patients.

Reasons prompting digitalis therapy in the acute care hospital.

BACKGROUND: The choice of administering digitalis to older patients with congestive heart failure (CHF) cannot be made on the account of univocally defined criteria because of uncertainty about efficacy and concern about safety of digitalis in this population. The purpose of this study was to verify whether the clinical characteristics on admission to the acute care hospital determine the use of digitalis therapy in elderly patients. METHODS: A total of 1239 patients (mean age 77.8 +/- 7.1 years, range 65-100 years, males 49.8%) consecutively admitted to 69 General Medicine and Geriatrics wards over a 4-month period were grouped by combining two dichotomous factors (Carlson's score > 4: definite or possible diagnosis of CHF; Carlson's score < 5: unlikely diagnosis of CHF; in-hospital adoption of digitalis therapy: yes or no) as follows: Group A: Carlson's score > 4, digitalis (n = 413); Group B: Carlson's score > 4, no digitalis (n = 260); Group C: Carlson's score < 5, digitalis (n = 104); Group D: Carlson's score < 5, no digitalis (n = 462). Variables significantly distinguishing groups were entered into a discriminant analysis aimed at assessing the group specificity of individual clinical profiles. RESULTS: Use of digoxin at home, atrial fibrillation, older age, and comorbidity (mainly COPD and chronic renal failure) characterized most of the patients given digoxin with or without a definite diagnosis of CHF. Clinical profiles of groups A, B, and C largely overlapped. CONCLUSION: Age, historical use of digitalis, and comorbidity might lead to seemingly incongruous digitalis prescription. The choice of adopting digitalis therapy cannot be reliably predicted on the basis of clinical variables only. Presently unexplored physician-related factors, such as cultural background, likely outweigh clinical variables in prompting digitalis prescription.

Trends in antihypertensive drugs in the elderly: the decline of thiazides.

The last decade has seen the publication of different editions of guidelines for the pharmacological treatment of hypertension that were based on the results of large, randomised trials. Since these guidelines were meant to inform practitioners, we analysed the pattern of prescription of antihypertensive agents between 1988 and 1997 among older hospitalised adults. Because of the wealth of data supporting the use of thiazides diuretics, we focused on diuretic prescription, to identify independent predictors of their utilisation. To this end, we used the GIFA database that includes patients admitted to academic medical centres throughout Italy between 1988 and 1997. We studied 5061 patients over 65 years of age selected among a population of 28 411, based on the diagnosis of arterial hypertension at discharge. The use of ACE-inhibitors has been raising steadily through the years, and they are the agents most commonly used since 1996. Calcium channel blockers showed a similar trend and were the top prescribing drug until 1995; afterwards, the documentation of potentially severe side effects has resulted in a nearly 20% reduction of their use. Beta-blockers have remained unpopular throughout the decade. Instead, the prescription of diuretics as a class showed a biphasic trend; an initial decrease with a prolonged steady state and a more recent raise. However, at a separate analysis, it was a evident that a progressive increase of the use of loop diuretics since 1988 has been paralleled by a nearly 50% reduction of thiazides prescriptions. Loop diuretics were more likely to be prescribed to older individuals, those with cardiac heart failure, coronary heart disease and high creatinine level. In contrast, independent predictors of thiazides use were female gender, good functional status, preserved renal function, and absence of cardiovascular comorbidity. In conclusion, despite continued recommendations to use thiazides diuretics for the treatment of hypertension among older individuals, their use has been declining steadily between 1988 and 1997. A possible explanation is that the choice to prescribe a thiazides diuretic is influenced by age, functional status and comorbidity.

Predicting length of stay of older patients with exacerbated chronic obstructive pulmonary disease.

Our study objective was to identify factors predicting length of hospital stay of older patients with exacerbated chronic obstructive pulmonary disease (COPD) through a multicenter, cross-sectional, retrospective study. We examined 3789 patients aged 74.3+/-11.1 years (mean+/-SD), 66.1% males, consecutively hospitalized in 32 wards of General Medicine and 31 of Geriatrics in acute care hospitals for exacerbated COPD in 10 bimonthly periods between 1988 and 1997. On admission, patients underwent a structured assessment of demographic data, nutritional status, cognitive and physical functions, comorbidity, and pharmacological therapy in the two weeks prior to admission. Patients were grouped according to whether their length of stay exceeded or not the 75th percentile of stay distribution in each bimonthly period. Variables univariately distinguishing groups were entered into a logistic regression analysis having long-stay as the dependent variable. Living alone (Odds Ratio 1.33, 95% Confidence Limits 1.03-1.70), use of more than 3 drugs prior to admission (OR 1.29, CL 1.09-1.51), use of drugs with respiratory depressant properties prior to admission (OR 1.24, CL 1.05-1.46), and the presence of more than 3 comorbid diseases (OR 1.88, CL 1.61-2.19) were independent correlates of long-stay. Age did not predict length of stay. In conclusion, selected health outcomes and indicators of disease severity, but not age, target COPD patients at risk of long-stay. Research is needed to verify whether these data can help program interventions aimed at shortening length of stay and, thus, at reducing annual hospitalization costs of the elderly.

Cognitive dysfunction as a major determinant of disability in patients with heart failure: results from a multicentre survey. On behalf of the GIFA (SIGG-ONLUS) Investigators.

Cognitive dysfunction is a frequent finding among older patients with left ventricular systolic dysfunction; however, the clinical outcomes of such a finding are unknown. Also, disability is a common condition in heart failure, poorly responding to commonly used cardiovascular medications. The association between cognitive dysfunction and disability was assessed in 1583 patients with heart failure, but without cerebrovascular disease, previous stroke, or Alzheimer's disease, who were enrolled during 2 years of a multicentre pharmacoepidemiology survey. The association between groups of variables (demographics, comorbid conditions, medications, and objective tests, including the Hodkinson abbreviated mental test) and functional disability (as indicated by need for intensive assistance in at least one of Katz' activities of daily living) was first analysed using separate age and sex adjusted logistic regression models. Those variables, significant at a p<0.1 level in these models, were simultaneously entered into an age and sex adjusted summary regression model. Among 1583 patients suitable for analysis, cognitive dysfunction (as detected by abbreviated mental test score <7) was detected in 265/461 disabled patients, and in 150/1122 independent subjects (p<0.0001). According to logistic regression analysis, cognitive dysfunction was associated with disability (OR=6.49; 95% CI=4.39-9.59) after adjusting for potential confounders.Thus, cognitive dysfunction in patients with heart failure is independently associated with disability, which currently represents an overwhelming medical and financial problem to patients, caregivers, and public health services. As early recognition and treatment of low cardiac output states might reverse cognitive dysfunction, cost effective treatment for heart failure should include systematic diagnostic and therapeutic approaches to cognitive dysfunction.

Impact of a new assessment system, the MDS-HC, on function and hospitalization of homebound older people: a controlled clinical trial.

OBJECTIVES: To evaluate the impact of a new assessment system, the Minimum Data Set for Home Care (MDS-HC), on the functional status and hospitalization rates of frail, community-dwelling older people. DESIGN: Single-blind randomized trial with 1-year follow-up. SETTING: Bergamo, Italy. PARTICIPANTS: All 187 subjects who were eligible for home care services delivered by two Health Districts between September 1998 and April 1999. INTERVENTION: Random allocation to an intervention group undergoing MDS-HC assessment or to a control group receiving conventional geriatric assessment with Barthel, Lawton and Brody, and Mini-Mental State Examination (MMSE) scales. MEASUREMENTS: Hospitalization, health services use and costs, and variations in functional status. RESULTS: Survival analysis indicated that the intervention group was admitted to the hospital later and less often than were controls (relative risk = 0.49, 95% confidence interval = 0.56-0.97). Health services were used to the same extent, but intervention subjects used more in-home help services. Total costs for the intervention group were 21% lower than for the control group. The adjusted mean scores of the activities of daily living index (51.7+/-36.1 vs 46.3+/-33.7; P = .05) and MMSE (19.9+/-8.9 vs 19.2+/-10.7; P = .03) were significantly improved in the intervention group as compared with the control group. CONCLUSIONS: The MDS-HC assessment instrument may provide a cost-saving approach to reducing institutionalization and functional decline in older people living in the community.

Serum magnesium levels and cognitive impairment in hospitalized hypertensive patients.

We performed this cross-sectional case control study to investigate the association between low serum magnesium levels and cognitive impairment in hypertensive hospitalized patients. The study was carried out in general medical care units at 81 hospitals participating in the Gruppo Italiano di Farmacovigilanza nell'Anziano (GIFA) study throughout Italy, and a total of 1058 patients with diagnoses of hypertension at the discharge were studied. The occurrence of cognitive impairment at discharge was the primary outcome of our study. Cognitive impairment was ascertained using the Hodkinson Abbreviated Mental Test (HAMT). Sociodemographic variables, body mass index, laboratory parameters, comorbidity, use of antihypertensive drugs and number of drugs were considered as potential confounders. Twenty-nine percent of the selected hypertensive patients were classified as having cognitive impairment. In univariate analysis, older age, female sex, and low educational level showed a significant trend for association to cognitive impairment. Moreover the proportion of subjects with cognitive impairment decreased with increasing alcohol consumption, and the prevalence of ex smokers and smokers was significantly lower in patients with cognitive impairment. The lower tertiles for serum albumin and creatinine clearance were more frequent among patients with cognitive impairment, and the lower tertile for serum magnesium levels was significantly more frequent in these patients. Number of drugs was slightly lower in cognitively impaired patients, while number of diagnoses and length of hospital stay were higher in these subjects. In the multivariate logistic regression analysis cognitive impairment decreased with increasing education level (highest education: OR 0.11; 95 per cent CI 0.05-0.25). The lower tertile for serum albumin (< 3.5 g/dl) was significantly associated to cognitive impairment (OR 2.14; 95 per cent CI 1.31-3.49), as well as the lower tertiles for serum magnesium (0.74-0.86 mmol/L: OR 1.54; 95 per cent CI 1.06-2.22; < 0.74 mmol/L: OR 1.75; 95 per cent CI 1.13-2.72]. Our results demonstrate the existence of a significant association between magnesium imbalance and cognitive impairment. These data suggest that the assessment of magnesium status may be of some relevance in hypertensive subjects with cognitive disorders.

Body mass index and mortality among hospitalized patients.

BACKGROUND: Body mass index (weight in kilograms divided by the square of the height in meters [BMI]) is known to be associated with overall mortality. However, the effect of age on excess mortality from all causes associated with obesity is controversial. The aim of the present study is to determine the effect of age on the relationship between BMI and mortality. METHODS: We analyzed data from a large collaborative observational study group, the Italian Group of Pharmacoepidemiology in the Elderly (GIFA), that collected data on hospitalized patients. A total of 18,316 patients consecutively admitted to 79 clinical centers during 5 different surveys in 1998, 1991, 1993, 1995, and 1997 were enrolled in the present study. The main outcome measure was the relative hazard ratio of death for different levels of BMI. RESULTS: Mortality rate was lowest among men and women with BMIs from 25.0 through 27.4 kg/m(2) (relative risk, 0.24; 95% confidence interval, 0.15-0.38). The graphed relationship between BMI and mortality in younger patients was hyperbolic, with increased death rates at the lowest and highest BMI rankings. On the contrary, the older patients showed an increased death rate at the lowest BMIs with only a slight elevation at the highest BMIs (>35 kg/m(2)). CONCLUSIONS: Our results suggest that BMI, a simple anthropometric measure of nutritional status, is an important predictor of mortality among young and old hospitalized patients. Even when controlling for clinical and functional variables, a low BMI remained a significant and independent predictor of shortened survival. Furthermore, the finding of the high BMI associated with minimum hazard in elderly subjects supports some past findings and opposes others and, if confirmed, has important implications for geriatric clinical guidelines.

Older age and in-hospital development of hypokalemia from loop diuretics: results from a multicenter survey. GIFA Investigators. Multicenter Italian Pharmacoepidemiologic Study Group.

BACKGROUND: Hypokalemia is a common finding among older patients taking diuretic medications. However, it is not known whether older age per se carries an increased risk of hypokalemia, particularly during a patient's treatment with loop diuretics. METHODS: The association between age and incident hypokalemia was examined in 18,872 patients with normal baseline serum potassium enrolled during three yearly multicenter surveys; 4,035 patients started receiving loop diuretics during their hospital stay. Demographic variables, comorbid conditions, medications, and objective tests that were associated with incident hypokalemia in separate age- and sex-adjusted logistic regression models were examined as potential confounders. RESULTS: Among patients with normal baseline serum potassium, the factors of age, presence of coronary disease or diabetes, comorbidity, the use of ACE inhibitors, loop diuretics, digitalis, corticosteroids, or insulin, and baseline serum potassium were associated with incident hypokalemia in initial models. After these variables were adjusted for, age (for each decade, odds ratio = 1.30; 95% confidence interval = 1.17-1.46; p < .0001) was associated with incident hypokalemia. The use of parenteral (2.30; 1.53-3.46; p < .0001) but not oral (1.16; 0.79-1.69; p = .44) loop diuretics was associated with hypokalemia. Eventually, age was associated with hypokalemia when the summary regression model was analyzed in patients taking loop diuretics (1.33; 1.03-1.71; p = .027), as well as in those taking intravenous loop diuretics only (1.84; 1.25-2.70; p = .002). CONCLUSIONS: Older age is independently associated with the in-hospital development of hypokalemia, particularly among patients taking loop diuretics. Monitoring of serum potassium levels is therefore advisable when older patients are treated with these agents.

Correlates of delayed referral for the diagnosis of dementia in an outpatient population.

BACKGROUND: Early diagnosis of dementia is critical, but there is usually a time lag between onset of symptoms and referral for neuropsychological testing and dementia diagnosis. We aimed to identify factors correlated with this delayed referral. METHODS: We studied 140 patients with cognitive deterioration referred to the Memory Clinic of the Catholic University (Rome) between 1995 and 1996. Alzheimer's disease or multi-infarct dementia was diagnosed according to National Institute of Neurological and Communicative Disorders and Stroke-Alzheimer's Disease and Related Disorders Association (NINCDS-ADRDA) criteria and Hachinski ischemic score. Global cognitive and neuropsychological functions were assessed with the Mini-Mental State Exam (MMSE) and the Mental Deterioration Battery. The performance on the Activities of Daily Living was used to measure physical function. The time between onset of signs of cognitive deterioration and referral for diagnosis (time to diagnosis: TTD) was estimated through a semistructured interview of the caregiver. Independent correlates of TTD were identified after adjustment for potential confounders and stratifying patients based on level of physical function. RESULTS: Of 127 eligible patients, 63% had Alzheimer's disease, 26% multi-infarct dementia, and 11% had dementia of other types. Mean age was 73.9 +/- 8.2 years, and 59% of patients were females. The mean TTD was 13.8 +/- 10.8 months and did not differ by gender, household composition, or type of dementia. For patients with normal physical function, increased age (beta = .50), female sex (beta = .51), and low MMSE score (beta = .36) were associated with longer TTD. Among patients with physical impairment, only MMSE score showed an association with TTD, but it was of opposite direction (beta = -.31). These associations were consistent by type of dementia. CONCLUSIONS: Age, gender, and degree of cognitive impairment are important correlates of the time between onset of signs/symptoms and referral for dementia diagnosis. These factors are independent of the type of dementia but are influenced by the level of physical function.