A comprehensive assessment of health care utilization among homeless adults under a system of universal health insurance

OBJECTIVES

We comprehensively assessed health care utilization in a population-based sample of homeless adults and matched controls under a universal health insurance system.

METHODS

We assessed health care utilization by 1165 homeless single men and women and adults in families and their age- and gender-matched low-income controls in Toronto, Ontario, from 2005 to 2009, using repeated-measures general linear models to calculate risk ratios and 95% confidence intervals (CIs).

RESULTS

Homeless participants had mean rates of 9.1 ambulatory care encounters (maximum = 141.1), 2.0 emergency department (ED) encounters (maximum = 104.9), 0.2 medical-surgical hospitalizations (maximum = 14.9), and 0.1 psychiatric hospitalizations per person-year (maximum = 4.8). Rate ratios for homeless participants compared with matched controls were 1.76 (95% CI = 1.58, 1.96) for ambulatory care encounters, 8.48 (95% CI = 6.72, 10.70) for ED encounters, 4.22 (95% CI = 2.99, 5.94) for medical-surgical hospitalizations, and 9.27 (95% CI = 4.42, 19.43) for psychiatric hospitalizations.

CONCLUSIONS

In a universal health insurance system, homeless people had substantially higher rates of ED and hospital use than general population controls; these rates were largely driven by a subset of homeless persons with extremely high-intensity usage of health services.

Predictors of medical or surgical and psychiatric hospitalizations among a population-based cohort of homeless adults

OBJECTIVES

We identified factors associated with inpatient hospitalizations among a population-based cohort of homeless adults in Toronto, Ontario.

METHODS

We recruited participants from shelters and meal programs. We then linked them to administrative databases to capture hospital admissions during the study (2005-2009). We used logistic regression to identify predictors of medical or surgical and psychiatric hospitalizations.

RESULTS

Among 1165 homeless adults, 20% had a medical or surgical hospitalization, and 12% had a psychiatric hospitalization during the study. These individuals had a total of 921 hospitalizations, of which 548 were medical or surgical and 373 were psychiatric. Independent predictors of medical or surgical hospitalization included birth in Canada, having a primary care provider, higher perceived external health locus of control, and lower health status. Independent predictors of psychiatric hospitalization included being a current smoker, having a recent mental health problem, and having a lower perceived internal health locus of control. Being accompanied by a partner or dependent children was protective for hospitalization.

CONCLUSIONS

Health care need was a strong predictor of medical or surgical and psychiatric hospitalizations. Some hospitalizations among homeless adults were potentially avoidable, whereas others represented an unavoidable use of health services.

High utilizers of emergency health services in a population-based cohort of homeless adults

OBJECTIVES

We identified predictors of emergency department (ED) use among a population-based prospective cohort of homeless adults in Toronto, Ontario.

METHODS

We assessed ED visit rates using administrative data from the Institute for Clinical Evaluative Sciences (2005-2009). We then used logistic regression to identify predictors of ED use. Frequent users were defined as participants with rates in the top decile (≥ 4.7 visits per person-year).

RESULTS

Among 1165 homeless adults, 892 (77%) had at least 1 ED visit during the study. The average rate of ED visits was 2.0 visits per person-year, whereas frequent users averaged 12.1 visits per person-year. Frequent users accounted for 10% of the sample but contributed more than 60% of visits. Predictors of frequent use in adjusted analyses included birth in Canada, higher monthly income, lower health status, perceived unmet mental health needs, and perceived external health locus of control from powerful others; being accompanied by a partner or dependent children had a protective effect on frequent use.

CONCLUSIONS

Among homeless adults with universal health insurance, a small subgroup accounted for the majority of visits to emergency services. Frequent use was driven by multiple predisposing, enabling, and need factors.

Validity of physician billing claims to identify deceased organ donors in large healthcare databases

Objective

We evaluated the validity of physician billing claims to identify deceased organ donors in large provincial healthcare databases.

Methods

We conducted a population-based retrospective validation study of all deceased donors in Ontario, Canada from 2006 to 2011 (n = 988). We included all registered deaths during the same period (n = 458,074). Our main outcome measures included sensitivity, specificity, positive predictive value, and negative predictive value of various algorithms consisting of physician billing claims to identify deceased organ donors and organ-specific donors compared to a reference standard of medical chart abstraction.

Results

The best performing algorithm consisted of any one of 10 different physician billing claims. This algorithm had a sensitivity of 75.4% (95% CI: 72.6% to 78.0%) and a positive predictive value of 77.4% (95% CI: 74.7% to 80.0%) for the identification of deceased organ donors. As expected, specificity and negative predictive value were near 100%. The number of organ donors identified by the algorithm each year was similar to the expected value, and this included the pre-validation period (1991 to 2005). Algorithms to identify organ–specific donors performed poorly (e.g. sensitivity ranged from 0% for small intestine to 67% for heart; positive predictive values ranged from 0% for small intestine to 37% for heart).

Interpretation

Primary data abstraction to identify deceased organ donors should be used whenever possible, particularly for the detection of organ-specific donations. The limitations of physician billing claims should be considered whenever they are used.

Drowning and the influence of hot weather

BACKGROUND

Drowning deaths are devastating and preventable. Public perception does not regard hot weather as a common scenario for drowning deaths. The objective of our study was to test the association between hot weather and drowning risk.

MATERIALS AND METHODS

We conducted a retrospective case-crossover analysis of all unintentional drowning deaths in Ontario, Canada from 1999 to 2009. Demographic data were obtained from the Office of the Chief Coroner. Weather data were obtained from Environment Canada. We used the pair-matched analytic approach for the case-crossover design to contrast the weather on the date of the drowning with the weather at the same location one week prior (control period).

RESULTS

We identified 1243 drowning deaths. The mean age was 40 years, 82% were male, and most events (71%) occurred in open water. The pair-matched analytic approach indicated that temperatures exceeding 30°C were associated with a 69% increase in the risk of outdoor drowning (OR = 1.69, 95% CI 1.23-2.25, p = 0.001). For indoor drowning, however, temperatures exceeding 30°C were not associated with a statistically significant increase in the risk of drowning (OR = 1.50, 95% CI 0.53-4.21, p = 0.442). Adult men were specifically prone to drown in hot weather (OR 1.67, 95% CI 1.19-2.34, p = 0.003) yet an apparent increase in risk extended to both genders and all age groups.

CONCLUSION

Contrary to popular belief, hot weather rather than cold stormy weather increases the risk of drowning. An awareness of this risk might encourage greater use of drowning prevention strategies known to save lives.

The exposure-crossover design is a new method for studying sustained changes in recurrent events

OBJECTIVES

To introduce a new design that explores how an acute exposure might lead to a sustained change in the risk of a recurrent outcome.

STUDY DESIGN AND SETTING

The exposure-crossover design uses self-matching to control within-person confounding due to genetics, personality, and all other stable patient characteristics. The design is demonstrated using population-based individual-level health data from Ontario, Canada, for three separate medical conditions (n > 100,000 for each) related to the risk of a motor vehicle crash (total outcomes, >2,000 for each).

RESULTS

The exposure-crossover design yields numerical risk estimates during the baseline interval before an intervention, the induction interval immediately ahead of the intervention, and the subsequent interval after the intervention. Accompanying graphs summarize results, provide an intuitive display to readers, and show risk comparisons (absolute and relative). Self-matching increases statistical efficiency, reduces selection bias, and yields quantitative analyses. The design has potential limitations related to confounding, artifacts, pragmatics, survivor bias, statistical models, potential misunderstandings, and serendipity.

CONCLUSION

The exposure-crossover design may help in exploring selected questions in epidemiology science.

JURaSSiC: accuracy of clinician vs risk score prediction of ischemic stroke outcomes

OBJECTIVE

We compared the accuracy of clinicians and a risk score (iScore) to predict observed outcomes following an acute ischemic stroke.

METHODS

The JURaSSiC (Clinician JUdgment vs Risk Score to predict Stroke outComes) study assigned 111 clinicians with expertise in acute stroke care to predict the probability of outcomes of 5 ischemic stroke case scenarios. Cases (n = 1,415) were selected as being representative of the 10 most common clinical presentations from a pool of more than 12,000 stroke patients admitted to 12 stroke centers. The primary outcome was prediction of death or disability (modified Rankin Scale [mRS] ≥3) at discharge within the 95% confidence interval (CI) of observed outcomes. Secondary outcomes included 30-day mortality and death or institutionalization at discharge.

RESULTS

Clinicians made 1,661 predictions with overall accuracy of 16.9% for death or disability at discharge, 46.9% for 30-day mortality, and 33.1% for death or institutionalization at discharge. In contrast, 90% of the iScore-based estimates were within the 95% CI of observed outcomes. Nearly half (n = 53 of 111; 48%) of participants were unable to accurately predict the probability of the primary outcome in any of the 5 rated cases. Less than 1% (n = 1) provided accurate predictions in 4 of the 5 cases and none accurately predicted all 5 case outcomes. In multivariable analyses, the presence of patient characteristics associated with poor outcomes (mRS ≥3 or death) in previous studies (older age, high NIH Stroke Scale score, and nonlacunar subtype) were associated with more accurate clinician predictions of death at 30 days (odds ratio [OR] 2.40, 95% CI 1.57-3.67) and with a trend for more accurate predictions of death or disability at discharge (OR 1.85, 95% CI 0.99-3.46).

CONCLUSIONS

Clinicians with expertise in stroke performed poorly compared to a validated tool in predicting the outcomes of patients with an acute ischemic stroke. Use of the risk stroke outcome tool may be superior for decision-making following an acute ischemic stroke.

Organ donation after death in Ontario: a population-based cohort study

BACKGROUND

Shortfalls in deceased organ donation lead to shortages of solid organs available for transplantation. We assessed rates of deceased organ donation and compared hospitals that had clinical services for transplant recipients (transplant hospitals) to those that did not (general hospitals).

METHODS

We conducted a population-based cohort analysis involving patients who died from traumatic brain injury, subarachnoid hemorrhage, intracerebral hemorrhage or other catastrophic neurologic conditions in Ontario, Canada, between Apr. 1, 1994, and Mar. 31, 2011. We distinguished between acute care hospitals with and without transplant services. The primary outcome was actual organ donation determined through the physician database for organ procurement procedures.

RESULTS

Overall, 87,129 patients died from catastrophic neurologic conditions during the study period, of whom 1930 became actual donors. Our primary analysis excluded patients from small hospitals, reducing the total to 79,746 patients, of whom 1898 became actual donors. Patients who died in transplant hospitals had a distribution of demographic characteristics similar to that of patients who died in other large general hospitals. Transplant hospitals had an actual donor rate per 100 deaths that was about 4 times the donor rate at large general hospitals (5.0 v. 1.4, p < 0.001). The relative reduction in donations at general hospitals was accentuated among older patients, persisted among patients who were the most eligible candidates and amounted to about 121 fewer actual donors per year (adjusted odds ratio 0.58, 95% confidence interval 0.36-0.92). Hospital volumes were only weakly correlated with actual organ donation rates.

INTERPRETATION

Optimizing organ donation requires greater attention to large general hospitals. These hospitals account for most of the potential donors and missed opportunities for deceased organ donation.

Opioid dose and risk of road trauma in Canada: a population-based study

BACKGROUND

Use of opioids may predispose drivers to road trauma, yet the effect of opioid dose on this association is unknown.

METHODS

We conducted a population-based nested case-control study of patients aged 18 to 64 years who received at least 1 publicly funded prescription for an opioid from April 1, 2003, through March 31, 2011. Cases were defined as having an emergency department visit related to road trauma. Patients without road trauma served as a control group matched to cases by age, sex, index year, prior road trauma, and a disease risk index. We compared the risk of road trauma among patients treated with doses of opioids ranging from very low to very high (<20 to ≥200 morphine equivalents daily). In a subgroup analysis, we stratified our analysis by driver status.

RESULTS

Among 549 878 eligible adults, we identified 5300 cases with road trauma and matched an equal number of controls. Multivariate adjustment yielded no significant association between escalating opioid dose and odds of road trauma (adjusted odds ratio ranged between 1.00 and 1.09). However, a significant association between opioid dose and road trauma was observed among drivers. Compared with very low opioid doses, drivers prescribed low doses had a 21% increased odds of road trauma (adjusted odds ratio, 1.21 [95% CI, 1.02-1.42]); those prescribed moderate doses, 29% increased odds (1.29 [1.06-1.57]); those prescribed high doses, 42% increased odds (1.42 [1.15-1.76]); and those prescribed very high doses, 23% increased odds (1.23 [1.02-1.49]).

CONCLUSIONS

Among drivers prescribed opioids, a significant relationship exists between drug dose and risk of road trauma. This association is distinct and does not appear with passengers, pedestrians, and others injured in road trauma.

Modern medicine is neglecting road traffic crashes

Don Redelmeier and Barry McLellan admonish the medical community for failure to act on the vast problem of road traffic crashes. Please see later in the article for the Editors' Summary

Urban emergency department overcrowding: defining the problem and eliminating misconceptions

STUDY OBJECTIVE

To develop an operational definition and a parsimonious list of postulated determinants for urban emergency department (ED) overcrowding.

METHODS

A panel was formed from clinical and administrative experts in pre-hospital, ED and hospital domains. Key studies and reports were reviewed in advance by panel members, an experienced health services researcher facilitated the panel's discussions, and a formal content analysis of audiotaped recordings was conducted.

RESULTS

The panel considered community, patient, ED and hospital determinants of overcrowding. Of 46 factors postulated in the literature, 21 were not retained by the experts as potentially important determinants of overcrowding. Factors not retained included access to primary care services and seasonal influenza outbreaks. Key determinants retained included admitted patients awaiting beds and patient characteristics. Ambulance diversion was considered to be an appropriate operational definition and proxy measure of ED overcrowding.

CONCLUSION

These results help to clarify the conceptual framework around ED overcrowding, and may provide a guide for future research. The relative importance of the determinants must be assessed by prospective studies.

Physicians' warnings for unfit drivers and the risk of trauma from road crashes

BACKGROUND

Physicians' warnings to patients who are potentially unfit to drive are a medical intervention intended to prevent trauma from motor vehicle crashes. We assessed the association between medical warnings and the risk of subsequent road crashes.

METHODS

We identified consecutive patients who received a medical warning in Ontario, Canada, between April 1, 2006, and December 31, 2009, from a physician who judged them to be potentially unfit to drive. We excluded patients who were younger than 18 years of age, who were not residents of Ontario, or who lacked valid health-card numbers under universal health insurance. We analyzed emergency department visits for road crashes during a baseline interval before the warning and a subsequent interval after the warning.

RESULTS

A total of 100,075 patients received a medical warning from a total of 6098 physicians. During the 3-year baseline interval, there were 1430 road crashes in which the patient was a driver and presented to the emergency department, as compared with 273 road crashes during the 1-year subsequent interval, representing a reduction of approximately 45% in the annual rate of crashes per 1000 patients after the warning (4.76 vs. 2.73, P<0.001). The lower rate was observed across patients with diverse characteristics. No significant change was observed in subsequent crashes in which patients were pedestrians or passengers. Medical warnings were associated with an increase in subsequent emergency department visits for depression and a decrease in return visits to the responsible physician.

CONCLUSIONS

Physicians' warnings to patients who are potentially unfit to drive may contribute to a decrease in subsequent trauma from road crashes, yet they may also exacerbate mood disorders and compromise the doctor-patient relationship. (Funded by the Canada Research Chairs program and others.).

Judging whether a patient is actually improving: more pitfalls from the science of human perception

Fallible human judgment may lead clinicians to make mistakes when assessing whether a patient is improving following treatment. This article provides a narrative review of selected studies in psychology that describe errors that potentially apply when a physician assesses a patient's response to treatment. Comprehension may be distorted by subjective preconceptions (lack of double blinding). Recall may fail through memory lapses (unwanted forgetfulness) and tacit assumptions (automatic imputation). Evaluations may be further compromised due to the effects of random chance (regression to the mean). Expression may be swayed by unjustified overconfidence following conformist groupthink (group polarization). An awareness of these five pitfalls may help clinicians avoid some errors in medical care when determining whether a patient is improving.

Social benefit payments and acute injury among low-income mothers

BACKGROUND

Human error due to risky behaviour is a common and important contributor to acute injury related to poverty. We studied whether social benefit payments mitigate or exacerbate risky behaviours that lead to emergency visits for acute injury among low-income mothers with dependent children.

METHODS

We analyzed total emergency department visits throughout Ontario to identify women between 15 and 55 years of age who were mothers of children younger than 18 years, who were living in the lowest socio-economic quintile and who presented with acute injury. We used universal health care databases to evaluate emergency department visits during specific days on which social benefit payments were made (child benefit distribution) relative to visits on control days over a 7-year interval (1 April 2003 to 31 March 2010).

RESULTS

A total of 153 377 emergency department visits met the inclusion criteria. We observed fewer emergencies per day on child benefit payment days than on control days (56.4 v. 60.1, p = 0.008). The difference was primarily explained by lower values among mothers age 35 years or younger (relative reduction 7.29%, 95% confidence interval [CI] 1.69% to 12.88%), those living in urban areas (relative reduction 7.07%, 95% CI 3.05% to 11.10%) and those treated at community hospitals (relative reduction 6.83%, 95% CI 2.46% to 11.19%). No significant differences were observed for the 7 days immediately before or the 7 days immediately after the child benefit payment.

INTERPRETATION

Contrary to political commentary, we found that small reductions in relative poverty mitigated, rather than exacerbated, risky behaviours that contribute to acute injury among low-income mothers with dependent children.

Simpson's paradox and the association between vitamin D deficiency and increased heart disease

Several recent investigations have highlighted a potential link between vitamin D deficiency and increased heart disease. Observational studies suggest cardioprotective benefits related to supplementation, but randomized trials remain to be conducted. This report adds a caution based on a statistical paradox that is rarely mentioned in formal medical training or in common medical journals. Insight into this phenomenon, termed Simpson's paradox, may prevent clinicians from drawing faulty conclusions about vitamin D deficiency and heart disease.

Roadway crash risks in recent immigrants

Adult immigrants are sometimes characterized as unsafe drivers and responsible for excess road crashes. We analyzed Canada's largest and most ethnically diverse province to assess whether recent immigrants had an increased risk of involvement as drivers in serious motor vehicle crashes. Overall, the study included 4,238,222 individuals followed for a median duration of 8.0 years. In total, 10,975 individuals were subsequently admitted to hospital as drivers involved in a crash, with a rate per 100,000 significantly lower among recent immigrants compared to long-term residents (158 vs 289, p<0.001). This difference was equal to a 45% relative reduction in the incidence of a crash (odds ratio=0.55, 95% confidence interval 0.52-0.58), persisted after adjustment for baseline characteristics (hazard ratio=0.61, 95% confidence interval 0.58-0.65), extended to extremes of crash severity, and was accentuated during initial years following immigration. These findings suggest that, contrary to popular opinion, recent immigrants are less prone to be drivers in serious motor vehicle crashes.