Estimating mean hospital cost as a function of length of stay and patient characteristics

Statistical models have been used to assess the influence of clinical and demographic factors on hospital charge and length of stay (LOS). Hospital costs constitute a significant proportion of overall expenditure in health care. With escalating costs, knowing the correlates of LOS and in-hospital cost is important for decisions on allocating resources. However, hospital charge and LOS are correlated. We describe two regression models that permit estimation of mean charges as a function of patient hospital stay and adjust for the influence of patient characteristics and treatment procedures on LOS and charge. In the first model, the mean charge over a specified duration is a weighted average of the expected cumulative charge, with weighting determined by the distribution of LOS. The second model for LOS and charge explicitly accounts for their correlation and yields estimates of the average charge per average LOS. The methods are applied to assess mean charges and mean charge per day by cardiac procedure in a cohort of patients hospitalized for acute myocardial infarction, while adjusting for the impact of patient demographic and clinical factors on LOS and charge. For relatively short hospital stays, and when only total hospital charges are available, these models provide a flexible approach to estimating summary measures on resource use while controlling for the effects of covariates on LOS and charge.     

The effects of air pollution on length of hospital stay for adult patients with asthma

Length of hospital stay (LOS) of asthma can be a reflection of the disease burden faced by patients, and it is also sensitive to air pollution. This study aims at estimating and validating the effects of air pollution and readmission on the LOS for those who have asthma, considering their readmission history, minimum temperature, and threshold effects of air pollutants. In addition, sex, age, and season were also constructed for stratification to achieve more precise and specific results. The results show that no significant effects of PM_2.5 and NO₂ on LOS were observed in any of the patients, but there were significant effects of PM_2.5 and NO₂ on LOS when a stratifying subgroup analysis was performed. The effect of PM₁₀ on LOS was found to be lower than that of PM_2.5 and higher than that of NO₂. SO₂ did not have a significant effect on LOS for patients with asthma in our study. Our study confirmed that the adverse effects of air pollutants (such as PM₁₀) on LOS for patients with asthma existed; in addition, these effects vary for different stratifications. We measured the effects of air pollutants on the LOS for patients with asthma, and this study offers policy makers quantitative evidence that can support relevant policies for health care resource management and ambient air pollutants control.     

Analysis of hospital length of stay and discharge destination using hazard functions with unmeasured heterogeneity

The hospital length-of-stay and the discharge destination of a Medicare patient are the outcomes of one decision process involving the interests of the patient, the hospital, and the firms offering covered post-hospital care. We use a competing risk hazard estimation procedure and adjust for unobserved heterogeneity with a non-parametric technique to identify significant factors in the decision process. A patient's health and socio-economic characteristics, the availability of informal care, local market area conditions, and Medicare policies influence length-of-stay and discharge destination. The substitution we find between hospital and post-hospital care and among post-hospital care alternatives has policy implications for Medicare.     

Nutrition status and risk factors associated with length of hospital stay for surgical patients

Objective: To diagnose the nutrition status of hospitalized patients and identify the risk factors associated with hospital length of stay (LOS). Methods: The subjective approach and the body mass index (BMI) were used to classify the nutrition status, and other indicators (anthropometry, biochemistry, and energy intake) were analyzed regarding their association with length of hospital stay of 350 patients. The chi‐square test was used to compare proportions, and the Mann‐Whitney or Kruskal‐Wallis test was used to compare continuous measures. Linear association was verified using Spearman's rank correlation coefficient. Cox's regression model was used to investigate factors associated with LOS. Results: Disease was the factor that influenced LOS the most in the studied population. Longer LOS prevailed in males (P < .0001), patients aged ≥60 years (P = .0008), patients with neoplasms (P < .0001), patients who lost weight during their hospital stay (P < .0001), and malnourished patients (P = .0034). There was a negative and significant, but weak, correlation between LOS and nutrition indicators (calf circumference, arm circumference, triceps skinfold thickness, subscapular skinfold thickness, arm fat area, lymphocyte count, and hemoglobin). Among adults, well‐nourished patients were 3 times more likely to be discharged sooner (P = .0002, RR = 3.3 [1.7–6.2]) than those who had some degree of malnutrition. Well‐nourished patients with digestive tract diseases (DTD) were also discharged sooner than malnourished patients with the same condition (P = .02, RR = 2.5 [1.1–5.8]). In patients with neoplasms, arm circumference was an independent risk factor to assess LOS (P = .009, RR = 1.1 [1.0–1.1]). Conclusions: LOS was associated with disease and nutrition status. Among the more common diseases, nutrition status according to the subjective approach determined the LOS for patients with DTD and nutrition status according to arm circumference determined the LOS for patients with neoplasms.     

医院感染对患者住院费用和住院时间的影响

目的分析医院感染对患者住院费用和住院时间的影响。方法对某院呼吸和血液内科患者,采用1∶1配对方法,比较医院感染者(感染组)与未感染者(对照组)之间住院费用和住院时间的差别。结果感染组住院费用平均高出对照组6 512.42元/人(P<0.05),平均住院日高出8.47 d(P<0.01)。结论医院感染导致患者住院费用增加和住院时间延长。     

老年营养风险指数与老年癌症及非癌症患者住院时间的关系研究

目的 应用老年营养风险指数(GNRI)评估患者的营养风险,探讨其与老年癌症及非癌症患者住院时间的关系。方法 选择老年住院患者37 267例,将其分为癌症组和非癌症组,应用GNRI评估患者入院时的营养风险;以患者死亡及出院为观察终点,住院时间(d)作为临床结局指标,采用边际结构模型探讨GNRI与老年癌症及非癌症患者住院时间的关联性。结果 超过一半(56.3%)的老年住院患者具有不同程度的营养风险;与非癌症患者相比,癌症患者GNRI水平(91.0±10.2)及无营养风险患者的比例(26.8%)更低,而具有低(24.6%)、中(29.0%)、高(19.7%)营养风险患者的比例均更高,差异均具有统计学意义(P＜0.05)。在控制其他混杂因素后,边际结构模型分析结果显示在癌症和非癌症患者中,住院时间均随营养风险程度的升高而延长,具有高营养风险的癌症患者住院时间最长,高达19.1(95% CI:17.5~20.8)d;在不同的营养风险分组中,癌症患者住院时间(14.5~19.1 d)均高于非癌症患者(10.1~15.2 d),差异均具有统计学意义(P＜0.05)。结论 GNRI 适用于老年癌症及非癌症患者的营养风险筛查评估,由GNRI评估的营养风险越高,患者住院时间越长。     

The relationship between the Balanced Budget Act and length of stay for Medicare patients in US hospitals

Background  The Balanced Budget Act (BBA) of 1997 and Balanced Budget Refinement Act (BBRA) of 1999 led to deep financial cuts for hospitals and nursing homes. Objectives  We examine the effects of these acts on hospital length of stay (LOS) for Medicare recipients. Methods  Using data for all short-stay community hospitals in the country, we compared LOS for Medicare patients before and after the BBA/BBRA relative to known determinants of LOS, e.g., hospital ownership, region, beds, financial performance, and conversion/change in ownership type. Results  Hospital LOS was reduced as a result of the acts. Reductions were more apparent for larger urban hospitals that provided safety-net services. LOS varied slightly by hospital ownership. Conclusion  This study is among the first to evaluate the impact of BBA and BBRA on hospital services. These acts had a negative effect on the ability of hospitals to continue offering safety-net services and negatively affected LOS.      

相位角预测手术病人营养风险及住院时间

目的:探讨生物电阻抗相位角(phase angle,PA)与胸外手术病人营养风险、住院时间(hospital length of stay,LOS)的关系。方法:60例胸部手术病人,应用营养风险筛查2002(NRS2002)、主观全面评定(Subjective Global Assessment,SGA)、白蛋白、人体成分分析、PA评价其营养风险及LOS。结果 :与对照组比较,手术病人PA偏低,男(6.0±1.0)vs(3.85±1.0),女(5.4±0.9)vs(4.9±0.6),P0.01。与正常PA比较,低PA者营养风险相对危险度,用NRS2002。无风险(RR 2.8,95%CI=1.2~6.9),中度风险(RR 3.9,95%CI=1.8~8.6),重度风险(RR4.2,95%CI=2.0~8.7);用SGA。营养良好(RR 2.5,95%CI=0.9~6.9),中度营养不良(RR 4.4,95%CI=2.1~9.4),重度营养不良(RR 3.9,95%CI=1.9~8.0),与NRS2002相似;与正常PA比较,低PA者住院时间倾向于延长(LOS≥21 d,RR=4.4,95%CI=2.2~8.8)。结论 :低PA与手术病人营养风险、LOS延长密切关联。PA测量有助于快速明确病人的营养风险对于确定病人营养干预和判断疾病转归提供了客观依据。     

The costs of caring for stroke patients in a GP-led community hospital: an application of programme budgeting and marginal analysis

There has been little evaluation of the role of community hospitals in the provision of integrated health care services in a primary care-led health system. The aim of this study was to model the probable changes in the use of NHS resources from the introduction of integrated stroke care in a general pracititioner-led community hospital. A programme budgeting and marginal analysis (PBMA) exercise was conducted combining practice data for the 'before' period and data from the literature to model the 'after' period. Data were collected from all patients discharged with a primary diagnosis of stroke 1994-96 in Nairn and Ardersier Total Fundholding pilot site, Highland Health Board, Scotland. Under several assumptions, a policy of early discharge of patients to the community hospital, and/or avoiding admission at the acute trust and admitting patients to the community hospital directly (except emergencies), is likely to result in a reduction of the total annual costs of treating stroke patients, from 183,000 pounds per annum to, at most, 74,000 pounds. The analysis of routine discharge data since integrated stroke care was set up has shown that progress has been made in shifting the treatment of patients from the acute trust to the community hospital. The care of stroke patients in a GP-led community hospital is likely to reduce the use of scarce health service resources. Current evidence suggests that health outcomes are unchanged due to early discharge, but further research is required to ensure that patients' health status and quality of life are maintained before such a policy is widely adopted.     

Recalibrated estimates of non-bacteremic and bacteremic pneumococcal community acquired pneumonia in hospitalized Canadian adults from 2010 to 2017 with addition of an extended spectrum serotype-specific urine antigen detection assay

Objective(s)In the context of age- and risk-based pneumococcal vaccine recommendations in Canada, this study presents updated data from active surveillance of pneumococcal community acquired pneumonia (pCAP) and invasive pneumococcal disease (IPD) in hospitalized adults from 2010 to 2017.MethodsS. pneumoniae was detected using culture (blood and sputum), and urine antigen detection (UAD). Serotyping was performed with Quellung, PCR, or using the PCV13- and PPV23 (non-PCV13)-specific UADs. Laboratory results, demographic, and outcome data were categorized by age (16–49, 50–64, and 65 + ) and by disease [non-bacteremic pCAP, bacteremic pCAP, and IPD(non-CAP)].Results11,129 CAP cases and 216 cases of IPD (non-CAP) were identified. Laboratory testing for S. pneumoniae was performed in 8912 CAP cases, identifying 1264 (14.2%) as pCAP. Of pCAP cases, 811 (64.1%) were non-bacteremic and 455 (35.9%) were bacteremic. Adults 65 + years represented 54.5% of non-bacteremic pCAP, 41.4% of bacteremic pCAP, and 48.6% of IPD cases. Adults 50–64 years contributed 30.3%, 33.1%, and 29.9%, respectively. In pCAP, PCV13 serotypes declined between 2010 and 2014 due to declines in serotypes 7F and 19A, then plateaued from 2015 to 2017 with persistence of serotype 3. In later study years, non-bacteremic pCAP was predominant, and PPV23 (non-PCV13) serotypes increased from 2015 to 2017, with serotypes 22F, 11A, and 9 N being most frequently identified. Compared to non-pCAP, pCAP cases were more likely to be admitted to intensive care units and require mechanical ventilation. These outcomes and mortality were more common in bacteremic pCAP and IPD, versus non-bacteremic pCAP.Conclusion(s)Along with IPD, pCAP surveillance (bacteremic and non-bacteremic) is important as their trends may differ over time. With insufficient herd protection from PCV13 childhood immunization, or use of PPV23 in adults, this study supports direct adult immunization with PCV13 or higher valency conjugate vaccines to reduce the residual burden of pCAP and IPD.