Sample size and statistical power in the hierarchical analysis of variance: applications in morphometry of the nervous system.

Analysis of variance is commonly used in morphometry in order to ascertain differences in parameters between several populations. Failure to detect significant differences between populations (type II error) may be due to suboptimal sampling and lead to erroneous conclusions; the concept of statistical power allows one to avoid such failures by means of an adequate sampling. Several examples are given in the morphometry of the nervous system, showing the use of the power of a hierarchical analysis of variance test for the choice of appropriate sample and subsample sizes. In the first case chosen, neuronal densities in the human visual cortex, we find the number of observations to be of little effect. For dendritic spine densities in the visual cortex of mice and humans, the effect is somewhat larger. A substantial effect is shown in our last example, dendritic segmental lengths in monkey lateral geniculate nucleus. It is in the nature of the hierarchical model that sample size is always more important than subsample size. The relative weight to be attributed to subsample size thus depends on the relative magnitude of the between observations variance compared to the between individuals variance.

Le médecin face aux inégalités sociales de santé: quel pouvoir d'action [General practitioners facing social inequalities in health: which power to act?].

The link between social inequalities and health has been known for many years, as attested by Villermé's work on the "mental and physical status of the working class" (1840). We have more and more insight into the nature of this relationship, which embraces not only material deprivation, but also psychological mechanisms related to social and interpersonal problems. Defining our possible role as physicians to fight against these inequalities has become a public health priority. Instruments and leads, which are now available to help us in our daily practice, are presented here.

Maximal aerobic power during running and cycling in obese and non-obese children.

The maximal aerobic capacity while running and cycling was measured in 22 prepubertal children (mean age +/- SD 9.5 +/- 0.8 years): 14 obese (47.3 +/- 10 kg) and 8 non-obese (31.1 +/- 6.1 kg). Oxygen consumption (VO2) and carbon dioxide production were measured by an open circuit method. Steady state VO2 was determined at different levels of exercise up to the maximal power on the cycloergometer (92 W in obese and 77 W in non-obese subjects) and up to the maximal running speed on the treadmill at a 2% slope (8.3 km/h in obese and 9.0 km/h in lean children). Expressed in absolute values, the VO2max in obese children was significantly higher than in controls (1.55 +/- 0.29 l/min versus 1.23 +/- 0.22 l/min, p < 0.05) for the treadmill test and comparable in the two groups (1.4 +/- 0.2 l/min versus 1.16 +/- 0.2 l/min, ns) for the cycloergometer test. When VO2max was expressed per kg fat free mass, the difference between the two groups disappeared for both tests. These data suggest that obese children had no limitation of maximal aerobic power. Therefore, the magnitude of the workload prescribed when a physical activity program is intended for the therapy of childhood obesity, it should be designed to increase caloric output rather than to improve cardiorespiratory fitness.

A program for the computation of power and determination of sample size in hierarchical experimental designs.

We have devised a program that allows computation of the power of F-test, and hence determination of appropriate sample and subsample sizes, in the context of the one-way hierarchical analysis of variance with fixed effects. The power at a fixed alternative is an increasing function of the sample size and of the subsample size. The program makes it easy to obtain the power of F-test for a range of values of sample and subsample sizes, and therefore the appropriate sizes based on a desired power. The program can be used for the 'ordinary' case of the one-way analysis of variance, as well as for hierarchical analysis of variance with two stages of sampling. Examples are given of the practical use of the program.

Exudative mineral losses after serious burns: a clue to the alterations of magnesium and phosphate metabolism.

Hypomagnesemia and hypophosphatemia are frequent after severe burns; however, increased urinary excretion does not sufficiently explain the magnitude of the mineral depletion. We measured the mineral content of cutaneous exudates during the first week after injury. Sixteen patients aged 34 +/- 9 y (mean +/- SD) with thermal burns were studied prospectively and divided in 3 groups according to the extent of their burn injury and the presence or absence of mineral supplements: group 1 (n = 5), burns covering 26 +/- 5% of body surface; group 2 (n = 6), burns covering 41 +/- 10%; and group 3 (n = 5), burns covering 42 +/- 6% with prescription of magnesium and phosphate supplements. Cutaneous exudates were extracted from the textiles (surgical drapes, dressings, sheets, etc) surrounding the patients from day 1 to day 7 after injury. Mean magnesium serum concentrations decreased below reference ranges in 12 patients between days 1 and 4 and normalized thereafter. Phosphate, normal on day 0, was low during the first week. Albumin concentrations, normal on day 0, decreased and remained low. Urinary magnesium and phosphate excretion were within reference ranges and not larger in group 3. Mean daily cutaneous losses were 16 mmol Mg/d and 11 mmol P/d (largest in group 2). Exudative magnesium losses were correlated with burn severity (r = 0.709, P = 0.003). Cutaneous magnesium losses were nearly four times larger than urinary losses whereas cutaneous phosphate losses were smaller than urinary phosphate losses. Mean daily losses of both magnesium and phosphate were more than the recommended dietary allowances. Exudative losses combined with urinary losses largely explained the increased mineral requirements after burn injury.

Five essays on bankruptcy

Chapter 2 Bankruptcy Initiation In The New Era of Chapter 11 2.1 Abstract The bankruptcy act of 1978 placed corporate managers (as debtor in possession) in control of the bankruptcy process. Between 2000 and 2001 managers apparently lost this control to secured creditors. This study examines financial ratios of firms filing for bankruptcy between 1993 and 2004 and tests the hypothesis that the change from manager to creditor control created or exacerbated the managerial (and dominant creditor) incentive to delay bankruptcy filing. We find a clear deterioration in the financial conditions of firms filing after 2001. This is consistent with managers (or creditors who control them) delaying filing for bankruptcy. We also observe patterns of operating losses and liquidations that suggest adverse economic consequences from such delay. Chapter 3 Bankruptcy Resolution: Priority of Claims with the Secured Creditor in Control 3.1 Abstract We present new evidence on the violation of priority of claims in bankruptcy using a sample of 222 firms that tiled for Chapter 11 bankruptcy over the 1993-2004 period. Our study reveals a dramatic reduction in the violations of priority of claims compared to research on prior periods. These results are consistent with changes in both court practices and laws transferring power to the secured creditors over our sample period. We also find an increase in the time from the date of a bankruptcy filing to reaching plan confirmation where priority is not violated. Chapter 4 Bankruptcy Resolution: Speed, APR Violations and Delaware 4.1 Abstract We analyze speed of bankruptcy resolution on a sample of 294 US firms filing for bankruptcy in the 1993-2004 period. We find strong association between type of Chapter II filing and speed of bankruptcy resolution. We also find that violations to the absolute priority rule reduce the time from bankruptcy filing to plan confirmation. This is consistent with the hypothesis that creditors are willing to grant concessions in exchange for faster bankruptcy resolution. Furthermore, after controlling for the type of filing and violations to the absolute priority rule, we do not find any difference in the duration of the bankruptcy process for firms filing in Delaware, New York, or other bankruptcy districts. Chapter 5 Financial Distress and Corporate Control 5.1 Abstract We examine the replacement rates of directors and executives in 63 firms filing for bank ruptcy during the 1995-2002 period. We find that over 76% of directors and executives are replaced in the four year period from the year prior to the bankruptcy filing through three years after. These rates are higher than those found in prior research and is consistent with changes in bankruptcy procedures and practice (i.e. the increased secured creditors control over the process due to both DIP financing and changes in the Uniform Commercial Code) having a significant impact on the corporate governance of firms in financial distress. Chapter 6 Financial Statement Restatements: Decision to File for Bankruptcy 6.1 Abstract On a sample of 201 firms that restated their financial statements we analyze the process of regaining investor trust in a two year period after the restatement. We find that 20% of firms that restate their financial statements tile for bankruptcy or restructure out of court. Our results also indicate that the decisions to change auditor or management is correlated with a higher probability of failure. Increased media attention appears to partly explain the decision of firms to restructure their debt or tile for bankruptcy.

Optimal dividend strategies for a compound Poisson risk process under transaction costs and power utility

We characterize the value function of maximizing the total discounted utility of dividend payments for a compound Poisson insurance risk model when strictly positive transaction costs are included, leading to an impulse control problem. We illustrate that well known simple strategies can be optimal in the case of exponential claim amounts. Finally we develop a numerical procedure to deal with general claim amount distributions.

Comparison of the power spectral changes of the voluntary surface electromyogram and M wave during intermittent maximal voluntary contractions.

INTRODUCTION: To compare the power spectral changes of the voluntary surface electromyogram (sEMG) and of the compound action potential (M wave) in the vastus medialis and vastus lateralis muscles during fatiguing contractions. METHODS: Interference sEMG and force were recorded during 48 intermittent 3-s isometric maximal voluntary contractions (MVC) from 13 young, healthy subjects. M waves and twitches were evoked using supramaximal femoral nerve stimulation between the successive MVCs. Mean frequency (F mean), and median frequency were calculated from the sEMG and M waves. Muscle fiber conduction velocity (MFCV) was computed by cross-correlation. RESULTS: The power spectral shift to lower frequencies was significantly greater for the voluntary sEMG than for the M waves (P < 0.05). Over the fatiguing protocol, the overall average decrease in MFCV (~25 %) was comparable to that of sEMG F mean (~22 %), but significantly greater than that of M-wave F mean (~9 %) (P < 0.001). The mean decline in MFCV was highly correlated with the mean decreases in both sEMG and M-wave F mean. CONCLUSIONS: The present findings indicated that, as fatigue progressed, central mechanisms could enhance the relative weight of the low-frequency components of the voluntary sEMG power spectrum, and/or the end-of-fiber (non-propagating) components could reduce the sensitivity of the M-wave spectrum to changes in conduction velocity.