Search tips
Search criteria

Results 1-22 (22)

Clipboard (0)

Select a Filter Below

Year of Publication
Document Types
1.  Assessing Multiple Medication Use With Probabilities of Benefits and Harms 
Journal of aging and health  2008;20(6):694-709.
A quantitative framework to assess harms and benefits of candidate medications in the context of drugs that a patient is already taking is proposed.
Probabilities of harms and benefits of a given medication are averaged to yield a utility value. The utility values of all medications under consideration are combined as a geometric mean to yield an overall measure of favorability. The grouping of medications yielding the highest favorability value is chosen.
Five examples of choosing between widely used candidate medications demonstrate the feasibility of the proposed framework.
The framework proposed provides a simple method for considering the trade-offs involved in prescribing multiple medications. It can be adapted to include additional parameters representing severity of condition, prioritization of outcomes, patient preferences, dosages, and medication interactions. Inconsistent reporting in the medical literature of data about benefits and harms of medications, dosages, and interactions constitutes its primary limitation.
PMCID: PMC3477770  PMID: 18625759
adverse effect; utility function; aging; trade-offs; multiple medications
2.  Association Between Indicators of Disability Burden and Subsequent Depression Among Older Persons 
Disability is associated with depression in older persons, yet the effect of disability burden on the likelihood of being depressed is uncertain.
A total of 754 community-living persons, aged ≥70, underwent monthly assessments in four essential activities of daily living and assessments of depression (yes/no) every 18 months for up to 108 months. Within each 18-month person-interval, participants’ disability burden was operationalized as none or any, and according to severity (none, mild, or severe) and chronicity (none, nonchronic, or chronic) given the highest level of severity or chronicity experienced during a given 18-month interval, respectively. A variable combining severity and chronicity (none, nonchronic mild, nonchronic severe, chronic–mild, or chronic–severe) was also created. Using generalized estimating equations, we evaluated the association between each indicator of disability burden and subsequent depression.
Participants who had any versus no disability during the previous 18 months were 65% more likely to experience subsequent depression (OR = 1.65; 95% confidence interval [CI] 1.34, 2.02). Quantifying severity (mild disability vs. none, OR = 1.43; 95% CI: 1.15, 1.79; severe disability vs. none, OR = 2.07; 95% CI 1.56, 2.74) and chronicity (nonchronic disability vs. none, OR = 1.44; 95% CI 1.13, 1.83; chronic disability vs. none, OR = 1.96; 95% CI 1.50, 2.55) indicated increasingly stronger associations with subsequent depression, with the highest likelihood of subsequent depression (OR = 2.42; 95% CI 1.78, 3.30) observed among participants with chronic–severe disability.
Quantifying the magnitude of disability burden, particularly on the basis of severity and chronicity, provides additional information regarding the likelihood of experiencing subsequent depression among older persons.
PMCID: PMC3605908  PMID: 22967459
Aging; Disability; Depression; Depressive symptoms; Prospective studies.
3.  Contribution of Individual Diseases to Death in Older Adults with Multiple Diseases 
To determine empirically the diseases contributing most commonly and strongly to death in older adults, accounting for coexisting diseases.
United States
Twenty two thousand eight hundred ninety Medicare Current Beneficiary Survey participants, a national representative sample of Medicare beneficiaries, enrolled during 2002 – 2006.
Chronic and acute diseases were ascertained from Medicare claims data. Diseases contributing to death during follow-up were identified empirically via regression models among all diseases with a frequency of ≥ 1% and hazard ratio for death of > 1. The additive contributions of these diseases, adjusting for co-existing diseases, were calculated using a longitudinal extension of average attributable fraction; 95% confidence intervals were estimated from bootstrapping.
Fifteen diseases and acute events contributed significantly to death, together accounting for nearly 70% of death. Heart failure (20.0%), dementia (13.6%), chronic lower respiratory disease (12.4%), and pneumonia (5.3%) made the largest contributions to death. Cancers, including lung, colorectal, lymphoma, and head and neck, together contributed to 5.6% of death. The other disease and events included acute kidney injury, stroke, septicemia, liver disease, myocardial infarction, and unintentional injuries.
The extent of the contribution of some diseases such as dementia and respiratory disease to death in older adults may be underappreciated, while the contribution of other diseases may be overestimated, with methods that focus on determining a single underlying cause. Current conceptualization of a single underlying cause may not account adequately for the contribution to death of coexisting diseases experienced by older adults.
PMCID: PMC3419332  PMID: 22734792
death; coexisting diseases; multiple chronic conditions
5.  Multivariate Graphical Methods Provide an Insightful Way to Formulate Explanatory Hypotheses from Limited Categorical Data 
Journal of Clinical Epidemiology  2011;65(2):179-188.
Graphical methods for generating explanatory hypotheses from limited categorical data are described and illustrated.
Study Design and Setting
Univariate, bivariate, multivariate, and multiplicative graphical methods were applied to clinical data regarding very ill older persons. The data to which these methods were applied were limited as to their nature (e.g., nominal categorical data) or quality (e.g., data subject to measurement error and missing values). Such limitations make confirmatory inference problematic but might still allow for meaningful generation of new explanatory hypotheses in some cases.
A striking feature of the graphical results from this study’s major illustrative application was that Post-Traumatic Stress Disorder (PTSD) following Intensive Care Unit (ICU) discharge occurred rarely and nearly always co-occurred with two or more other mental health conditions. These results suggest the explanatory hypothesis that PTSD in this context is less attributable to single traumatic causes than to acute illnesses contributing to a cascade of mental health decrements.
Illustrative applications of a sequence of graphical procedures yield more informative and less abstract representations of limited data than do descriptive statistics alone, and by doing so, they aid in the formulation of explanatory hypotheses.
PMCID: PMC3250573  PMID: 21889310
graphical methods; multivariate data; older patients; logic; critical care
6.  Bayesian Time-Series Analysis of a Repeated-Measures Poisson Outcome With Excess Zeroes 
American Journal of Epidemiology  2011;174(11):1230-1237.
In this article, the authors demonstrate a time-series analysis based on a hierarchical Bayesian model of a Poisson outcome with an excessive number of zeroes. The motivating example for this analysis comes from the intensive care unit (ICU) of an urban university teaching hospital (New Haven, Connecticut, 2002–2004). Studies of medication use among older patients in the ICU are complicated by statistical factors such as an excessive number of zero doses, periodicity, and within-person autocorrelation. Whereas time-series techniques adjust for autocorrelation and periodicity in outcome measurements, Bayesian analysis provides greater precision for small samples and the flexibility to conduct posterior predictive simulations. By applying elements of time-series analysis within both frequentist and Bayesian frameworks, the authors evaluate differences in shift-based dosing of medication in a medical ICU. From a small sample and with adjustment for excess zeroes, linear trend, autocorrelation, and clinical covariates, both frequentist and Bayesian models provide evidence of a significant association between a specific nursing shift and dosing level of a sedative medication. Furthermore, the posterior distributions from a Bayesian random-effects Poisson model permit posterior predictive simulations of related results that are potentially difficult to model.
PMCID: PMC3254157  PMID: 22025357
autocorrelation; Bayes theorem; models, statistical; periodicity; time series
7.  College Graduation Rates for Minority Students in a Selective Technical University: Will Participation in a Summer Bridge Program Contribute to Success? 
There are many approaches to solving the problem of underrepresentation of some racial and ethnic groups and women in scientific and technical disciplines. Here, the authors evaluate the association of a summer bridge program with the graduation rate of underrepresented minority (URM) students at a selective technical university. They demonstrate that this 5-week program prior to the fall of the 1st year contains elements reported as vital for successful student retention. Using multivariable survival analysis, they show that for URM students entering as fall-semester freshmen, relative to their nonparticipating peers, participation in this accelerated summer bridge program is associated with higher likelihood of graduation. The longitudinal panel data include more than 2,200 URM students.
PMCID: PMC3489281  PMID: 23136456
persistence; retention; underrepresented minority; survival analysis; STEM; bridge program
8.  An examination of effect estimation in factorial and standardly-tailored designs 
Many clinical trials are designed to test an intervention arm against a control arm wherein all subjects are equally eligible for all interventional components. Factorial designs have extended this to test multiple intervention components and their interactions. A newer design referred to as a ‘standardly-tailored’ design, is a multicomponent interventional trial that applies individual interventional components to modify risk factors identified a priori and tests whether health outcomes differ between treatment arms. Standardly-tailored designs do not require that all subjects be eligible for every interventional component. Although standardly-tailored designs yield an estimate for the net effect of the multicomponent intervention, it has not yet been shown if they permit separate, unbiased estimation of individual component effects. The ability to estimate the most potent interventional components has direct bearing on conducting second stage translational research.
We present statistical issues related to the estimation of individual component effects in trials of geriatric conditions using factorial and standardly-tailored designs. The medical community is interested in second stage translational research involving the transfer of results from a randomized clinical trial to a community setting. Before such research is undertaken, main effects and synergistic and or antagonistic interactions between them should be identified. Knowledge of the relative strength and direction of the effects of the individual components and their interactions facilitates the successful transfer of clinically significant findings and may potentially reduce the number of interventional components needed. Therefore the current inability of the standardly-tailored design to provide unbiased estimates of individual interventional components is a serious limitation in their applicability to second stage translational research.
We discuss estimation of individual component effects from the family of factorial designs and this limitation for standardly-tailored designs. We use the phrase ‘factorial designs’ to describe full-factorial designs and their derivatives including the fractional factorial, partial factorial, incomplete factorial and modified reciprocal designs. We suggest two potential directions for designing multicomponent interventions to facilitate unbiased estimates of individual interventional components.
Full factorial designs and their variants are the most common multicomponent trial design described in the literature and differ meaningfully from standardly-tailored designs. Factorial and standardly-tailored designs result in similar estimates of net effect with different levels of precision. Unbiased estimation of individual component effects from a standardly-tailored design will require new methodology.
Although clinically relevant in geriatrics, previous applications of standardly-tailored designs have not provided unbiased estimates of the effects of individual interventional components.
Future directions to estimate individual component effects from standardly-tailored designs include applying D-optimal designs and creating independent linear combinations of risk factors analogous to factor analysis.
Methods are needed to extract unbiased estimates of the effects of individual interventional components from standardly-tailored designs.
PMCID: PMC3477845  PMID: 18375650
9.  Effect of Dissemination of Evidence in Reducing Injuries from Falls 
The New England journal of medicine  2008;359(3):252-261.
Falling is a common and morbid condition among elderly persons. Effective strategies to prevent falls have been identified but are underutilized.
Using a nonrandomized design, we compared rates of injuries from falls in a region of Connecticut where clinicians had been exposed to interventions to change clinical practice (intervention region) and in a region where clinicians had not been exposed to such interventions (usual-care region). The interventions encouraged primary care clinicians and staff members involved in home care, outpatient rehabilitation, and senior centers to adopt effective risk assessments and strategies for the prevention of falls (e.g., medication reduction and balance and gait training). The outcomes were rates of serious fall-related injuries (hip and other fractures, head injuries, and joint dislocations) and fall-related use of medical services per 1000 person-years among persons who were 70 years of age or older. The interventions occurred from 2001 to 2004, and the evaluations took place from 2004 to 2006.
Before the interventions, the adjusted rates of serious fall-related injuries (per 1000 person-years) were 31.2 in the usual-care region and 31.9 in the intervention region. During the evaluation period, the adjusted rates were 31.4 and 28.6, respectively (adjusted rate ratio, 0.91; 95% Bayesian credibility interval, 0.88 to 0.94). Between the preintervention period and the evaluation period, the rate of fall-related use of medical services increased from 68.1 to 83.3 per 1000 person-years in the usual-care region and from 70.7 to 74.2 in the intervention region (adjusted rate ratio, 0.89; 95% credibility interval, 0.86 to 0.92). The percentages of clinicians who received intervention visits ranged from 62% (131 of 212 primary care offices) to 100% (26 of 26 home care agencies).
Dissemination of evidence about fall prevention, coupled with interventions to change clinical practice, may reduce fall-related injuries in elderly persons.
PMCID: PMC3472807  PMID: 18635430
We determined the association between clinically significant depressive symptoms, often referred to as depression, and subsequent transitions between no disability, mild disability, severe disability, and death.
Prospective cohort study.
General community in greater New Haven, Connecticut, from March 23, 1998, to December 31, 2008.
Seven-hundred fifty four persons, aged 70 years or older.
Monthly assessments of disability in essential activities of daily living and assessments of depressive symptoms every 18 months using a short-form of the Center for Epidemiologic Studies of Depression Scale for up to 129 months.
Depressed participants were more likely than those who were non-depressed to transition from a state of no disability to mild (HR= 1.52; 95% CI 1.25, 1.85) and severe disability (HR=1.57; 95% CI 1.22, 2.01), and from a state of mild disability to severe disability (HR=1.33; 95% CI 1.06, 1.65); and were less likely to transition from a state of mild disability to no disability (HR=0.69; 95% CI 0.57, 0.85) and from a state of severe disability to no disability (HR=0.50; 95% CI 0.31, 0.79).
Depressive symptoms are associated with transitions into and out of disabled states, and with increased likelihood of transitioning from mild to severe disability. More broadly, our findings underscore the complexity of the relationship between depressive symptoms and disability. Future work is needed to evaluate the likely reciprocal relationship between depression and functional transitions in older persons.
PMCID: PMC3164788  PMID: 21873834
Depression; Depressive symptoms; Disability; Prospective studies
Annals of Internal Medicine  2012;156(2):131-140.
Relatively little is known about why older persons develop long-term disability in community mobility.
To identify the risk factors and precipitants for long-term disability in walking ¼ mile and driving a car, respectively.
Prospective cohort study from March 1998 to December 2009.
Greater New Haven, Connecticut.
641 persons, 70+ years, who were active drivers or nondisabled in walking ¼ mile. Persons who were physically frail were oversampled.
Candidate risk factors were assessed every 18 months. Disability in community mobility and exposure to potential precipitants, which included illnesses/injuries leading to hospitalization or restricted activity, respectively, were assessed every month. Disability lasting ≥6 consecutive months was considered long term.
318 (56.0%) and 269 (53.1%) participants developed long-term disability in walking and driving, respectively. Seven risk factors were independently associated with walking disability, while eight were associated with driving disability; the strongest associations for each outcome were found for older age and lower score on the Short Physical Performance Battery. The effects of the precipitants on long-term disability were large, with multivariable hazard ratios for each outcome greater than 6 for hospitalization and 2.4 for restricted activity. The largest differences in absolute risk were generally observed for participants who had a specific risk factor and were subsequently hospitalized.
The observed associations may not be causal. The severity of precipitants was not assessed. The effect of the precipitants may have been underestimated because their exposure after the initial onset of disability was not evaluated.
Long-term disability in community mobility is common among older persons. Multiple risk factors, together with subsequent precipitants, greatly increase the likelihood of developing long-term mobility disability.
Primary Funding Source
National Institute on Aging.
PMCID: PMC3278794  PMID: 22250144
To determine the association between depression and functional recovery among community-living older persons who had a decline in function after an acute hospital admission.
Prospective cohort study.
General community in greater New Haven, Connecticut, from March 1998 to December 2008.
Seven-hundred fifty four persons, aged 70 years or older.
Hospitalization and disability in essential activities of daily living (ADLs) and mobility were assessed each month for up to 129 months, and depressive symptoms were assessed every 18 months using the Center for Epidemiologic Studies of Depression (CES-D) Scale. Functional recovery was defined as returning to the community within 6 months at or above the pre-hospital level of ADL function and mobility, respectively.
A decline in ADL function and mobility was observed following 42% and 41% of the hospitalizations, respectively. After controlling for several potential confounders, clinically significant depressive symptoms (CES-D ≥20) was associated with a lower likelihood of recovering mobility function (HR= 0.79; 95%CI 0.63, 0.98), but not ADL function (HR= 0.91; 95%CI 0.75, 1.10), within 6 months of hospitalization.
Following a disabling hospitalization among community-living older persons, those with pre-existing depression may be less likely to recover their pre-hospitalization level of mobility function, but not ADL function. Yet, the reasons remain to be elucidated.
PMCID: PMC3136548  PMID: 21668916
Depression; hospitalization; functional decline; recovery
13.  Factors associated with persistent delirium following ICU admission in an older medical patient population 
Journal of critical care  2010;25(3):540.e1-540.e7.
This study was designed to identify factors associated with persistent delirium in an older medical ICU population.
Materials and Methods
Prospective cohort study of 309 consecutive medical ICU patients age ≥60. Persistent delirium was defined as delirium occurring in the ICU and continuing upon discharge to the ward. The Confusion Assessment Method (CAM) was used to assess for delirium. Patient demographics, severity of illness, and medication data were collected. Univariate and multivariate analysis were used to assess factors associated with persistent delirium.
Of 309 consecutive admissions to the ICU, 173 patients had ICU delirium, survived the ICU stay, and provided ward data. One-hundred patients (58%) had persistent delirium. In a multivariable logistic regression model, factors significantly associated with persistent delirium included age >75 years (OR, 2.52, 95% CI, 1.23–5.16), opioid (morphine equivalent) dose >54 mg/day (OR, 2.90, 95% CI, 1.15–7.28), and haloperidol (OR, 2.88, 95% CI, 1.38–6.02); change in code status to ‘Do Not Resuscitate’ (DNR) (OR, 2.62, 95% CI 0.95–7.35) and dementia (OR, 1.93, 95% CI 0.95–3.93) had less precise associations.
Age, use of opioids and haloperidol were associated with persistent delirium. Further research is needed regarding the use of haloperidol and opioids on persistent delirium.
PMCID: PMC2939229  PMID: 20413252
persistent delirium; critical care; aged
14.  Gerontologic Biostatistics: The Statistical Challenges of Clinical Research with Older Study Participants 
The medical and personal circumstances of older persons present challenges for designing and analyzing clinical research studies in which they participate. These challenges presented by elderly study samples are not unique but they are sufficiently distinctive to warrant deliberate and systematic attention. Their distinctiveness originates in the multifactorial etiologies of geriatric health syndromes and the multiple morbidities accruing with aging at the end of life. The objective of this article is to identify a set of statistical challenges arising in research with older persons that should be considered conjointly in the practice of clinical research and that should be addressed systematically in the training of biostatisticians intending to work with gerontologists, geriatricians, and older study participants. The statistical challenges include design and analytical strategies for multicomponent interventions, multiple outcomes, state transition models, floor and ceiling effects, missing data, and mixed methods. The methodological and pedagogical themes of this article will be integrated by a description of a proposed subdiscipline of “gerontologic biostatistics” and supported by the introduction of new set of statistical resources for researchers working in this area. These conceptual and methodological resources have been developed in the context of several collaborating Claude D. Pepper Older Americans Independence Centers.
PMCID: PMC2918405  PMID: 20533963
clinical research; statistics; aging; study design
Disability among older persons is a complex and highly dynamic process, with high rates of recovery and frequent transitions between states of disability. The role of intervening illnesses and injuries (i.e. events) on these transitions is uncertain.
To evaluate the relationship between intervening events and transitions among states of no disability, mild disability, severe disability and death, and to determine the association of physical frailty with these transitions.
Design, Setting, and Participants
Prospective cohort study, conducted in greater New Haven, Connecticut, from March 1998 to December 2008, of 754 community-living persons, aged 70 years or older, who were nondisabled at baseline in four essential activities of daily living: bathing, dressing, walking, and transferring. Telephone interviews were completed monthly for more than 10 years to assess disability and ascertain exposure to intervening events, which included illnesses and injuries leading to either hospitalization or restricted activity. Physical frailty (defined as gait speed >10 seconds on the rapid gait test) was assessed every 18 months through 108 months.
Main Outcome Measure
Transitions between no disability, mild disability, and severe disability, and 3 transitions from each of these states to death, were evaluated each month.
Hospitalization was strongly associated with 8 of the 9 possible transitions, with increased multivariable hazard ratios (HR) as high as 168 (95% confidence interval [CI], 118–239) for the transition from no disability to severe disability and decreased HRs as low as 0.41 (95% CI, 0.30–0.54) for the transition from mild disability to no disability. Restricted activity also increased the likelihood of transitioning from no disability to both mild and severe disability (HR [CI]: 2.59 [2.23–3.02] and 8.03 [5.28–12.21]), respectively, and from mild disability to severe disability (1.45 [1.14–1.84]), but was not associated with recovery from mild or severe disability. For all nine of the transitions, the presence of physical frailty accentuated the associations of the intervening events. For example, the absolute risk of transitioning from no disability to mild disability within one month after hospitalization for frail individuals was 12.4% (95% CI, 12.1%–12.7%) vs 4.9% (4.7%–5.1%) for non-frail individuals. Among the possible reasons for hospitalization, fall-related injury conferred the highest likelihood of developing new or worsening disability.
Among older persons, particularly those who were physically frail, intervening illnesses and injuries greatly increased the likelihood of developing new or worsening disability. Only the most potent events, i.e. those leading to hospitalization, reduced the likelihood of recovery from disability.
PMCID: PMC3124926  PMID: 21045098
16.  Disability in activities of daily living, depression, and quality of life among older medical ICU survivors: a prospective cohort study 
Accurate measurement of quality of life in older ICU survivors is difficult but critical for understanding the long-term impact of our treatments. Activities of daily living (ADLs) are important components of functional status and more easily measured than quality of life (QOL). We sought to determine the cross-sectional associations between disability in ADLs and QOL as measured by version one of the Short Form 12-item Health Survey (SF-12) at both one month and one year post-ICU discharge.
Data was prospectively collected on 309 patients over age 60 admitted to the Yale-New Haven Hospital Medical ICU between 2002 and 2004. Among survivors an assessment of ADL's and QOL was performed at one month and one-year post-ICU discharge. The SF-12 was scored using the version one norm based scoring with 1990 population norms. Multivariable regression was used to adjust the association between ADLs and QOL for important covariates.
Our analysis of SF-12 data from 110 patients at one month post-ICU discharge showed that depression and ADL disability were associated with decreased QOL. Our model accounted for 17% of variability in SF12 physical scores (PCS) and 20% of variability in SF12 mental scores (MCS). The mean PCS of 37 was significantly lower than the population mean whereas the mean MCS score of 51 was similar to the population mean. At one year mean PCS scores improved and ADL disability was no longer significantly associated with QOL. Mortality was 17% (53 patients) at ICU discharge, 26% (79 patients) at hospital discharge, 33% (105 patients) at one month post ICU admission, and was 45% (138 patients) at one year post ICU discharge.
In our population of older ICU survivors, disability in ADLs was associated with reduced QOL as measured by the SF-12 at one month but not at one year. Although better markers of QOL in ICU survivors are needed, ADLs are a readily observable outcome. In the meantime, clinicians must try to offer realistic estimates of prognosis based on available data and resources are needed to assist ICU survivors with impaired ADLs who wish to maintain their independence. More aggressive diagnosis and treatment of depression in this population should also be explored as an intervention to improve quality of life.
PMCID: PMC3041645  PMID: 21294911
17.  Days of Delirium Are Associated with 1-Year Mortality in an Older Intensive Care Unit Population 
Rationale: Delirium is a frequent occurrence in older intensive care unit (ICU) patients, but the importance of the duration of delirium in contributing to adverse long-term outcomes is unclear.
Objectives: To examine the association of the number of days of ICU delirium with mortality in an older patient population.
Methods: We performed a prospective cohort study in a 14-bed ICU in an urban acute care hospital. The patient population comprised 304 consecutive admissions 60 years of age and older.
Measurements and Main Results: The main outcome was 1-year mortality after ICU admission. Patients were assessed daily for delirium with the Confusion Assessment Method for the ICU and a validated chart review method. The median duration of ICU delirium was 3 days (range, 1–46 d). During the follow-up period, 153 (50%) patients died. After adjusting for relevant covariates, including age, severity of illness, comorbid conditions, psychoactive medication use, and baseline cognitive and functional status, the number of days of ICU delirium was significantly associated with time to death within 1 year post-ICU admission (hazard ratio, 1.10; 95% confidence interval, 1.02–1.18).
Conclusions: Number of days of ICU delirium was associated with higher 1-year mortality after adjustment for relevant covariates in an older ICU population. Investigations should be undertaken to reduce the number of days of ICU delirium and to study the impact of this reduction on important health outcomes, including mortality and functional and cognitive status.
PMCID: PMC2784414  PMID: 19745202
delirium; aging; mortality; intensive care
18.  Does Gender Impact Intensity of Care Provided to Older Medical Intensive Care Unit Patients? 
Introduction. Women receive less aggressive critical care than men based on prior studies. No documented studies evaluate whether men and women are treated equally in the medical intensive care unit (MICU). The Therapeutic Intervention Scoring System-28 (TISS-28) has been used to examine gender differences in mixed ICU studies. However, it has not been used to evaluate equivalence of care in older MICU patients. We hypothesize that given nonsignificant, baseline health differences between genders at MICU admission, the level of care provided would be equivalent. Methods. Prospective cohort of 309 patients ≥60 years old in the MICU of an urban university teaching hospital. Explanatory variables were demographic data and baseline measures. Primary outcomes were TISS-28 scores and MICU interventions. We compare TISS-28 scores by gender using a statistical test of equivalence. Results. Women were older and had more chronic respiratory failure at MICU admission. Using equivalence limits of ±15% on gender-based scores of TISS-28, MICU interventions were equivalent. Supplementary analysis showed no statistically significant association between gender and mortality. Conclusions. In contrast with other reports from the cardiac critical care literature, as measured by the TISS-28, gender-based care delivered to older MICU patients in this cohort was equivalent.
PMCID: PMC2964007  PMID: 20981259
To identify risk factors for five different subtypes of disability.
Design, Setting and Participants
Prospective cohort study of 754 community-living residents of greater New Haven, Connecticut, who were 70 years or older and initially nondisabled in four essential activities of daily living (bathing, dressing, walking, and transferring).
Candidate risk factors were measured every 18 months for 90 months during comprehensive home-based assessments. Disability was assessed during monthly telephone interviews for up to 108 months. Among participants who were nondisabled at the start of an 18-month interval, incident episodes of five different disability subtypes were determined during the subsequent 18 months: transient, short-term, long-term, recurrent, and unstable.
The cumulative incidence rates (95% confidence intervals) per 100 person-intervals were 9.8 (8.9–10.6) for transient disability, 3.8 (3.3–4.3) for short-term disability, 7.1 (6.4–7.8) for long-term disability, 4.7 (4.1–5.3) for recurrent disability, and 4.4 (3.9–5.0) for unstable disability. In a multivariate analysis, the Short Physical Performance Battery (SPPB) was associated with each of the five disability subtypes, with adjusted hazard ratios ranging from 1.10 for transient disability to 1.35 for long-term disability. The only other factors associated with short-term, long-term, and recurrent disability were stroke, visual impairment, and poor grip strength, respectively. Transient disability and unstable disability shared the same set of risk factors—depressive symptoms, stroke, and poor grip strength—in addition to the SPPB.
Our results provide mixed evidence to support the distinct nature of the five disability subtypes.
PMCID: PMC2782909  PMID: 19694870
aged; cohort studies; risk factors; disability evaluation; activities of daily living
20.  Benzodiazepine and opioid use and the duration of ICU delirium in an older population 
Critical care medicine  2009;37(1):177-183.
There is a high prevalence of delirium in older medical intensive care unit (ICU) patients and delirium is associated with adverse outcomes. We need to identify modifiable risk factors for delirium in the ICU, such as medication use. The objective of this study was to examine the impact of benzodiazepine or opioid use on the duration of ICU delirium in an older medical population.
Prospective cohort study.
Fourteen-bed medical intensive care unit in an urban university teaching hospital.
304 consecutive admissions age 60 and older.
Main Outcome Measurements
The main outcome measure was duration of ICU delirium, specifically the first episode of ICU delirium. Patients were assessed daily for delirium with the Confusion Assessment Method for the ICU (CAM-ICU) and a validated chart review method. Our main predictor was the receipt of benzodiazepines or opioids during ICU stay. A multivariable model was developed using Poisson rate regression.
Delirium occurred in 239 of 304 patients (79%). The median duration of ICU delirium was 3 days with a range of 1-33 days. In a multivariable regression model receipt of a benzodiazepine or opioid (RR, 1.64, 95% CI, 1.27-2.10) was associated with increased delirium duration. Other variables associated with delirium duration in this analysis include preexisting dementia (RR, 1.19, 95% CI 1.07-1.33), receipt of haloperidol (RR, 1.35, 95% CI, 1.21-1.50), and severity of illness (RR, 1.01, 95% CI, 1.00-1.02).
The use of benzodiazepines or opioids in the ICU is associated with longer duration of a first episode of delirium. Receipt of these medications may represent modifiable risk factors for delirium. Clinicians caring for ICU patients should carefully evaluate the need for benzodiazepines, opioids and haloperidol.
PMCID: PMC2700732  PMID: 19050611
delirium; critical care; risk factors; aged; benzodiazepines; opioids; haloperidol
21.  The Use of Missingness Screens in Clinical Epidemiologic Research Has Implications for Regression Modeling 
Journal of clinical epidemiology  2007;60(12):1239-1245.
Properly handling missing data is a challenge, especially when working with older populations that have high levels of morbidity and mortality. Methods have been developed to understand whether missing values are ignorable. We illustrate their use in a study of intensive care unit (ICU) delirium in an older cohort.
Study Design
Little’s Missing Completely at Random “MCAR test” (1988) assesses whether values are missing completely at random. The “Index of Sensitivity to Nonignorability (ISNI)” by Troxel and colleagues (2004) assesses the extent to which values are missing at random. Use of such missingness screens introduces complications for regression modeling, and, particularly, for risk factor selection. We propose a model fitting process that incorporates the use of missingness screens, controls for collinearity, and selects variables based on model fit.
In a case study with simulated missing data, the proposed model fitting process identifies more actual risk factors for ICU delirium than does a complete case analysis.
Use of imputation and inverse weighting methods for handling missing data assist in the identification of risk factors. They do so accurately only when correct assumptions are made about the nature of missing data. Missingness screens enable researchers to investigate these assumptions.
PMCID: PMC2443713  PMID: 17998078
Missing data; regression modeling; delirium; intensive care unit; older adults
22.  A Method for Partitioning the Attributable Fraction of Multiple Time-Dependent Coexisting Risk Factors for an Adverse Health Outcome 
American journal of public health  2012;103(1):177-182.
We decomposed the total effect of coexisting diseases on a timed occurrence of an adverse outcome into additive effects from individual diseases.
In a cohort of older adults enrolled in the Precipitating Events Project in New Haven County, Connecticut, we assessed a longitudinal extension of the average attributable fraction method (LE-AAF) to estimate the additive and order-free contributions of multiple diseases to the timed occurrence of a health outcome, with right censoring, which may be useful when relationships among diseases are complex. We partitioned the contribution to death into additive LE-AAFs for multiple diseases.
The onset of heart failure and acute episodes of pneumonia during follow-up contributed the most to death, with the overall LE-AAFs equal to 13.0% and 12.1%, respectively. The contribution of preexisting diseases decreased over the years, with a trend of increasing contribution from new onset of diseases.
LE-AAF can be useful for determining the additive and order-free contribution of individual time-varying diseases to a time-to-event outcome.
PMCID: PMC3518339  PMID: 22515873

Results 1-22 (22)