Impact of psychiatric follow-up frequency on outcomes and waiting times.

OBJECTIVES: This study determined whether naturally occurring but significantly different outpatient follow-up frequencies are associated with clinical outcomes and service waiting times. STUDY DESIGN: Longitudinal retrospective study. METHODS: This study was conducted in an outpatient setting. Participants consisted of 340 patients with major depressive disorder who were randomly assigned to 4 psychiatrists and were followed at a variable frequency defined by the clinician. Patients were assessed at baseline and at every visit with structured interviews and self-reported questionnaires. These groups were also compared according to their baseline characteristics, treatment, and appointment frequencies. Little's law was used to estimate the impact of modifying the appointment frequencies on the service waiting time. RESULTS: The demographic variables, prescriptions, and depression severity at intake of patients across the 4 groups were similar. The mean times between appointments of the 4 groups were significantly different (87.0, 46.9, 67.9, and 61.5 days, respectively; P < .001), but these differences in outpatient follow-up frequency were not associated with clinical outcomes (eg, mean last Quick Inventory of Depressive Symptomatology Self-Report score, 10.5, 10.0, 11.9, and 9.7; P = .25). However, different outpatient follow-up frequencies had an estimated impact on waiting times for access to care; using Little's law, it was observed that the waiting list would be eliminated by reducing by 23.9% the follow-up frequencies of the 3 psychiatrists with the highest frequencies. CONCLUSIONS: Although variations in appointment frequencies do not appear to have a major impact on clinical outcomes, they could be managed to achieve significant improvements in the accessibility of the clinic.

Community Treatment Order Outcomes in Quebec: A Unique Jurisdiction.

OBJECTIVE: We study compulsory community treatment orders (CTOs) for patients with severe and persistent mental illness (SPMI). Focusing on a unique jurisdiction in Canada that allows for long duration CTOs with strict enforcement procedures, our objectives are to determine whether extended duration CTOs are effective and to determine whether associated hospitalization costs are reduced. METHOD: A mirror image, naturalistic design was employed using patients as their own controls to enhance external validity. No inclusive or exclusive criteria were employed for the 367 SPMI clinic patients who were studied over a 5-year period. Detailed documentation of the dates of all CTOs, long-acting antipsychotic injections (LAIs), emergency visits, hospitalizations, duration of hospitalizations, crimes and/or police involvement were collected. To study the relation between CTO and injection adherence, we use a mixed-effect linear regression model. To study the effect of injection adherence and hospitalization, we use survival analysis via Kaplan-Meier and Cox survival models. RESULTS: CTO and non-CTO patients did not differ with respect to demographics, but CTO patients were significantly more severely ill. Following a CTO, adherence to LAIs increased over time (P < 0.001). The average time the patients spent in the community, that is, outside the hospital, was significantly longer under a CTO, and the duration of hospitalizations was decreased. CONCLUSIONS: LAI adherence and outpatient office visits were enhanced by extended duration CTOs, as was time out of the hospital. The shorter duration of hospital stays implies cost savings. These must be weighed against their undesirable coercive nature.

Patient-centric design of long-term care networks.

Long-term care networks may soon buckle under the weight of overwhelming demand. We present two dynamic, large-scale mixed-integer programs for long-term care network design that execute jointly strategic and tactical facility location, modular capacity acquisition, and patient-assignment decisions. The first model is an adaptive network-design model whose focus is more strategic in nature, whereas the second model focuses exclusively on the expansion of an existing long-term care network and incorporates additional tactical decisions such as patient backlogs. Working directly with the president of the Order of Québec Nurses-the provincial organization representing over 75,000 nurses-we incorporate facets such as assignment permanence, as well as develop and measure patient-centric quality-of-life proxies such as geographic mis-assignment and un-assigned patients, the latter of which is quantified via parametric optimization. Various network-design and patient-assignment policies are explored. We conclude that the use of home care as an alternative to long-term care facilities is cost prohibitive under specific conditions. Employing a bisection algorithm, we identify the implicit cost placed on keeping medically stable elderly patients in a hospital ward, concluding no cost savings are generated from such a policy. The model is analyzed and validated using empirical data from the long-term care network in Montréal, Canada.

The impact of specialization of hospitals on patient access to care; a queuing analysis with an application to a neurological hospital.

We study the impact of specialization on the operational efficiency of a multi-hospital system. The mixed outcomes of recently increasing hospital mergers and system re-configuration initiatives have raised the importance of studying such organizational changes from all the relevant perspectives. We consider two configuration scenarios for a multi-hospital system. The first scenario assumes that all the hospitals in the system are general, which implies they can provide care to all types of patients. In the alternative configuration, we specialize each hospital in certain level of care, which means they serve only specific types of patients. By considering an extensive number of possible settings for a multi-hospital system, we characterize the situations in which one scenario outperforms the other in terms of extending access of patients to care. Our results show that whenever the percent of patients with shorter length of stay in the system increases, specialization of healthcare services can maximize the accessibility of care. Also, if the patient load is balanced between all hospitals in the system, it seems more likely that all hospitals benefit from specialization. We conclude that the strategic decision of designing a multi-hospital system requires careful consideration of patient mix among arrivals, relative length of stay of patients, and distribution of patient load between hospitals.

Transport Mode Selection for Toxic Gases: Rail or Road?

A significant majority of hazardous materials (hazmat) shipments are moved via the highway and railroad networks, wherein the latter mode is generally preferred for long distances. Although the characteristics of highway transportation make trucks the most dominant surface transportation mode, should it be preferred for hazmat whose accidental release can cause catastrophic consequences? We answer this question by first developing a novel and comprehensive assessment methodology-which incorporates the sequence of events leading to hazmat release from the derailed railcars and the resulting consequence-to measure rail transport risk, and second making use of the proposed assessment methodology to analyze hazmat transport risk resulting from meeting the demand for chlorine and ammonia in six distinct corridors in North America. We demonstrate that rail transport will reduce risk, irrespective of the risk measure and the transport corridor, and that every attempt must be made to use railroads to transport these shipments.

Predicting the need for CT imaging in children with minor head injury using an ensemble of Naive Bayes classifiers.

OBJECTIVE: Using an automatic data-driven approach, this paper develops a prediction model that achieves more balanced performance (in terms of sensitivity and specificity) than the Canadian Assessment of Tomography for Childhood Head Injury (CATCH) rule, when predicting the need for computed tomography (CT) imaging of children after a minor head injury. METHODS AND MATERIALS: CT is widely considered an effective tool for evaluating patients with minor head trauma who have potentially suffered serious intracranial injury. However, its use poses possible harmful effects, particularly for children, due to exposure to radiation. Safety concerns, along with issues of cost and practice variability, have led to calls for the development of effective methods to decide when CT imaging is needed. Clinical decision rules represent such methods and are normally derived from the analysis of large prospectively collected patient data sets. The CATCH rule was created by a group of Canadian pediatric emergency physicians to support the decision of referring children with minor head injury to CT imaging. The goal of the CATCH rule was to maximize the sensitivity of predictions of potential intracranial lesion while keeping specificity at a reasonable level. After extensive analysis of the CATCH data set, characterized by severe class imbalance, and after a thorough evaluation of several data mining methods, we derived an ensemble of multiple Naive Bayes classifiers as the prediction model for CT imaging decisions. RESULTS: In the first phase of the experiment we compared the proposed ensemble model to other ensemble models employing rule-, tree- and instance-based member classifiers. Our prediction model demonstrated the best performance in terms of AUC, G-mean and sensitivity measures. In the second phase, using a bootstrapping experiment similar to that reported by the CATCH investigators, we showed that the proposed ensemble model achieved a more balanced predictive performance than the CATCH rule with an average sensitivity of 82.8% and an average specificity of 74.4% (vs. 98.1% and 50.0% for the CATCH rule respectively). CONCLUSION: Automatically derived prediction models cannot replace a physician's acumen. However, they help establish reference performance indicators for the purpose of developing clinical decision rules so the trade-off between prediction sensitivity and specificity is better understood.

Improving post-stroke health outcomes: can facilitated care help?

OBJECTIVES: The objectives of this study were (1) identifying the patterns of post-stroke care, (2) determining the care-provider and patient characteristics associated with optimal management of post-stroke care and (3) estimating the potential influence of various facilitated care policies on outcomes. METHODOLOGY: The 3946 subjects included in the study were admitted to one of Quebec's acute-care hospitals with confirmed diagnosis of stroke and subsequently discharged to their home. The records related to fee-for-service billings of this sample were obtained for the 3 months following discharge and used to define the care-provider path for each stroke survivor. These paths were analyzed and the potential impact of various facilitated care interventions was estimated via a Markov model. RESULTS: The rate of mortality for this sample was 3.2% during the first 3 months after discharge. For the patients who were re-hospitalized, however, the mortality rates were up to 10.3% depending on the care-provider visited prior to re-hospitalization. Our analyses indicate that by avoiding such critical sub-paths via facilitated care, it is possible to achieve improvements in health outcomes as well as cost. DISCUSSION: There is a window of opportunity for improving community-based post-stroke care. Facilitated care policies concerning planned visits upon discharge from hospital or following ER visits can improve the outcomes.

The in-hospital interval: a description of EMT time spent in the emergency department.

OBJECTIVE: We conducted a time-motion study of emergency medical technician (EMT) flow in an urban, academic emergency department (ED). Our objective was to describe the activity of the EMTs during their time in the ED. Secondary objectives included the association of time of day, age, and triage code with the various time intervals. METHODS: In this descriptive study, we combined information from two databases: prospectively collected time-motion data of EMTs presenting to one ED and an electronically collected prehospital call database of time data. The pretriage, triage, and posttriage time intervals were calculated, as well as total time spent in the ED as a proportion of total call time. Mean times with 95% confidence intervals (CIs) were reported. Analysis of variance was performed to examine the associations of time of day, age, and triage code with time intervals. RESULTS: Data were available for 152 calls. The mean pretriage interval was 8.79 (95% CI, 7.55-10.04) minutes, the mean triage interval was 5.14 (95% CI, 4.49-5.79) minutes, and the mean posttriage interval was 31.33 (95% CI, 29.08-33.58) minutes. The proportion of the total call time that was spent in the ED was 45%. Subgroup analysis showed significant differences only between total time spent in the ED in the 7:30-10:00 AM period as compared with the other periods. CONCLUSIONS: More time was spent in the pretriage and posttriage intervals as compared with the triage interval. Further time-motion studies in the ED will be necessary to plan interventions aimed at decreasing the time spent in-hospital by EMTs.

On the significance of reducing the need for stroke patients to visit the emergency department.

We studied the care-provider paths followed by 3,946 patients in Quebec in 2001. We showed that the patients flow during the three months preceding discharge from hospital can be represented by a Markov model with memory. This model enables study of four major scenarios to improve health outcomes, workloads and cost efficiency in the overall community-based care delivery system. Based on the field data, we establish that increasing the availability of specialists, family physicians and general practitioners to mitigate the need for ER visits would be an effective strategy for improvement. A comprehensive policy to support stroke patients needs to incorporate both hospital-based and community-based care delivery processes. The seamless flow of patients through the healthcare providers in such an integrated system is crucial for achieving successful outcomes. Emergency rooms (ER) have a crucial role in this context, since in many cases ER acts as the hospital's "gate keeper", determining if a patient needs to be (re)admitted. In this paper, we establish (based on field data) that mitigating the ER visits of stroke patients improves health outcomes, distribution of workload across the healthcare system as well as associated costs. To this end, we make use of a Markov modeling framework, where the aggregate patient flow information is represented in a compact form through the use of a transition-probability matrix. This allows us to investigate the system-wide impact of several plausible scenarios with regards to the delivery of community-based care to stroke patients who are recently discharged from hospital.