Abstract
Background
Investigators are actively testing interventions intended to increase lifespan and wish to test whether the interventions increase maximum lifespan. Based on the fact that one cannot be assured of observing population maximum lifespans in finite samples, in previous work, we constructed and validated several tests of difference in the upper parts of lifespan distributions between a treatment group and a control group by testing whether the probabilities that observations are above some threshold defining 'old' or being in the tail of the survival distribution are equal in the two groups. However, a limitation of these tests is that they do not consider how much above the threshold any particular observation is.
Methods
In this article we propose new methods which improve upon our previous tests by considering not only whether an observation is above some threshold, but also the magnitudes by which observations exceed the threshold.
Results
Simulations show that the new methods control type I error rates quite well and that the power of the new methods is usually higher than that of the tests we previously proposed. In illustrative analyses of two real datasets involving rodents, when setting the threshold equal to 110 (100) weeks for the first (second) datasets, the new methods detected differences in 'maximum lifespan' between groups at nominal alpha levels of 0.01 (0.05) for the first (second) datasets and provided more significant results than competitor tests.
Conclusion
The new methods not only have good performance in controlling the type I error rates but also improve the power compared with the tests we previously proposed.
Background
Investigators are actively testing interventions intended to increase lifespan [1]. Caloric restriction (CR) is the intervention most well established as able to increase lifespan in experimental models [2], and investigators are now seeking other interventions that may mimic the lifeprolonging effects of CR without requiring a reduction in caloric intake [3]. It is frequently said that CR not only increases average lifespan, but also 'maximum' lifespan [4]. Many researchers in the field of aging therefore wish to test whether other interventions increase maximum lifespan.
Recognizing this and the fact that one cannot be assured of observing population maximum lifespans in finite samples, Wang et al. [5] constructed and validated several tests (hereafter, the 'WangAllison tests') of differences in the upper parts of lifespan distributions by building on the work of Redden et al. [6] in the area of quantile regression. Wang et al. also showed that a commonly used test for differences in maximum lifespan that involved comparing the means of the top p% (e.g., top 10%) of each of two samples (e.g., a treatment and a control sample) was not valid in that it had an excessive type1 error rate. Nevertheless, there is appeal to using the full continuity of information in the upper tails of the sample distribution, and colleagues have recently suggested to us that a limitation of the WangAllison tests is that they only treat individual lifespans as being above or below some threshold defining 'old' or being in the tail of the survival distribution. That is, the WangAllison tests do not consider how much above the threshold any particular observation is, only whether the observation is above the threshold. We acknowledge this limitation and in response, we herein develop new tests that utilize the continuity of information among observations that exceed the threshold of interest, are more powerful than competing tests, including the WangAllison tests, in most cases, and remain valid under the null hypothesis of no effect on 'maximum' lifespan.
Methods
Development of the tests
Consider an experiment with two groups, treatment and control. The extension to more than two groups is straightforward (see discussion section). Let X be an indicator variable taking the value 1 for observations in the treatment group and 0 for observations in the control group. Let Y denote survival time. Let τ denote some threshold chosen by the investigator to denote an extreme portion of the distribution. In survival studies, τ can be chosen in advance to correspond to an age considered 'old' (e.g., 30 months in mice) or set to some high sample percentile (e.g., the 90th). Critically important, τ must be set to the same value for the two groups. That is, if τ is to be defined by an upper sample quantile, it should be the upper sample quantile of both of the two groups combined, not of each group separately.
Although not described in exactly these terms in the paper by Wang et al. [5], the WangAllison tests essentially create a new variable, W, where for the i^{th }subject, W_{i }≡ 0 if Y_{i }≤ τ, and W_{i }≡ 1 if Y_{i }> τ, and subsequently tests whether W is associated with X using an appropriate test statistic.
Thus, the WangAllison tests test the following null hypothesis:
A problem with the WangAllison tests is that, hypothetically, P (Y > τX = 1) may equal P (Y > τX = 0) and yet the average magnitude by which lifespans exceed τ when X = 1 may be radically different than when X = 0. This is exemplified in the hypothetical frequency distributions depicted in Figure 1. Note that these hypothetical distributions are not intended to be realistic, but only to clarify the point.
Figure 1. The left graph is the density for control group (X = 0), 0.9*Weibull(5.73, 106.6)*I(X ≤ 130) + 0.1*Weibull(5.40, 100.06)*I(X > 130), and the right graph is the density for treatment group (X = 1), 0.9*Weibull(5.73, 106.6)*I(X ≤ 130) + 0.1*Weibull(5.45, 130.06)*I(X > 130), where P(Y > τX = 1) = P(Y > τX = 0) and yet the average magnitude by which lifespans exceed τ when X = 1 is different than when X = 0. τ is 90^{th }percentile of the all observations in treatment and control groups.
Let X^{1 }and X^{0 }denote the numbers of observations with Y_{i }> τ in the treatment group and control group, respectively. The WangAllison tests use the test procedures for two independent binomial proportions [7] and these procedures require that X^{1 }and X^{0 }are independent. In the WangAllison tests, if the threshold is set in advance according to prior knowledge, X^{1 }and X^{0 }can satisfy the requirement of independence. But if τ is set to be the 90th percentile, X^{1 }and X^{0 }may not be independent, this creates a theoretical problem. However, on an empirical level, our simulations show that in the sample sizes we considered, this is not an apparent problem because the WangAllison tests have very high power and can control type I error quit well in the simulation studies and are practical for the lifespan studies). When X^{1 }and X^{0 }are not independent, simulation studies (including estimation of power and type I error) are an effective way to evaluate the methods (such as WangAllison tests) using the test procedures for two independent binomial proportions.
An alternative to testing H_{0,A }is to test the following conceptually related but mathematically distinct null hypothesis:
where μ (•) denotes the population mean (or expectation) of (•). Though appealing, a problem with testing H_{0,B }is that when P (Y > τX = 1) >> P (Y > τX = 0) or P (Y > τX = 1) <<P (Y > τX = 0), for any finite sample with equal initial assignment to the two groups, E [n_{0}] <<E [n_{1}] or E [n_{0}] >> E [n_{1}], where E [n_{0}] denotes the expected number of observations in the control group for which Y > τ, and E [n_{1}] denotes the expected number of observations in the treatment group for which Y > τ. This imbalance between E [n_{0}] and E [n_{1}] will greatly reduce the power to reject H_{0,B}. In fact, in the extreme, when either P (Y > τX = 1) or P (Y > τX = 0), there will be zero power to reject H_{0,B }(actually, it is appropriate to say that H_{0,B }is undefined in such cases). Such a situation is exemplified in the hypothetical frequency distributions depicted in Figure 2. Again, these hypothetical distributions are not intended to be realistic, but only to clarify the point.
Figure 2. The left graph is the density for control group (X = 0), 0.9*Weibull(5.07, 93.52)*I(X ≤ 130) + 0.1*Weibull(5.40, 100.06)*I(X > 130), and the right graph for treatment group (X = 1), 0.6*Weibull(5.07, 93.52)*I(X ≤ 130) + 0.4*Weibull(5.40, 100.06)*I(X > 130), where P(Y > τX = 1) ≠ P(Y > τX = 0), μ (Y Y > τ ∩ X = 1) = μ (Y Y > τ ∩ X = 0), and μ (•) denotes the population mean of (•). τ is 90^{th }percentile of the all observations in treatment and control groups.
Thus, one can conceive situations in which the power to reject H_{0,A }will be zero and yet the upper tails of the distribution are clearly different. Similarly, one can conceive situations in which the power to reject H_{0,B }will be zero and yet again the upper tails of the distribution are clearly different. Hence, we propose a singlestep unionintersection test [8] of the following compound null hypothesis:
We construct the test of H_{0,C }with the following simple procedure. Define a new variable Z such that Z_{i }≡ I(Y_{i }> τ)Y_{i}, where I(•) denotes the indicator function taking on values of one if (•) is true and zero otherwise. One can then simply conduct an appropriate test (several candidates will be considered below) of whether the population mean of Z is significantly different between the treatment and control groups. This approach (hereafter new tests), has several desirable properties. First and foremost, when an appropriate test statistic is used, the approach will be valid. That is, unlike the conditional ttests (CTTs) commonly used and shown to be invalid by Wang et al. [5], when H_{0,C }is true, it will only be rejected 100*α% of the time at the nominal α level even if f(YY ≤ τ ∩ X = 1) ≠ f(YY ≤ τ ∩ X = 0), where f(•) denotes the probability density function of (•).
Note that expectation (or population mean) of Z, μ(Z) = P(Y > τ) μ(Y  Y > τ). Therefore the new test for H_{0,C }is really testing whether
while the method for H_{0,B }is testing whether μ (Y  Y > τ ∩ X = 1) = μ(Y  Y > τ ∩ X = 0) and the method for H_{0,A }is testing whether P(Y > τ  X = 1) = P(Y > τ  X = 0). The mean difference of μ(Z) between two groups consists of two components: the difference between probabilities P(Y > τ  X = 1) and P(Y > τ  X = 0) and the difference between expectations μ (Y  Y > τ ∩ X = 1) and μ (Y  Y > τ ∩ X = 0). The test for H_{0,A }focuses on the first component and the test for H_{0,A }focuses on the second one, while the test for H_{0,C }is related to both components.
We also note that Dominici and Zeger [9] studied similar mean difference components for two groups (cases and controls) by estimating the mean difference Δ(v) for the two groups conditional on a vector of covariates v for zeroinflated data through smooth quantile ratio estimation with regression,
where, Y is nonnegative random variable denoting the health expenditures. While Dominici and Zeger [9] estimate the mean difference of nonnegative random variables (Y) for two groups, our methods test the mean difference of random variables (Y) which are greater than threshold τ.
Evaluation of the tests
We evaluate the tests via computer simulation. For each scenario simulated, we evaluate the tests at the 2tailed .05 α level and at the 2tailed .01 α level using 5,000 simulated datasets per scenario (except for permutation tests where we use 1,000 datasets per scenario and 1,000 random permutations by Monte Carlo sampling for each dataset). In simulation 1, we first evaluate performance in simulation under the null hypothesis H_{0,C }(i.e., both H_{0,A }and H_{0,B }are true) and yet f (YY ≤ τ ∩ X = 1) is radically different from f (YY ≤ τ ∩ X = 0). After showing that the tests remain valid even in these extreme circumstances, we compare their power in several scenarios (simulations 2–4) described below. For each scenario, we assumed that there were two groups with an equal number of subjects per group. We ran scenarios with 50, 80, or 100 subjects in each of the two groups, realistic sample sizes for animal model longevity research.
We simulated data using a concatenation of Weibull distributions to flexibly emulate the data observed in a real study [10] of obese animals (control; X = 0) versus animals that were obese and then lost weight via CR (treatment; X = 1). Specifically, For example, in simulations 1–4, we simulated Y from the following distribution:
where j = 0 to 1, lifespan (Y) is measured in weeks, a_{j,L }and b_{i,L }are the parameters of a Weibull distribution for the lower 90% of the distribution, and a_{j,U }and b_{i,U }are the parameters of a Weibull distribution for the upper 10% of the distribution. r_{j }is a proportion parameter, for example r_{j }= 0.9. The specific values of the parameters used are provided in Figure 3.
Figure 3. Parameter values and distributions for component Weibull distributions used in each simulation.
Delineation of tests to be evaluated
Each of the tests listed below was implemented in two manners, first with τ set in advance to a fixed lifespan value (130 weeks), and second with τ set at the sample 90^{th }percentile of the two groups combined. In reallife situations, one usually does know the threshold of interest a priori. We do recognize that we will not have such knowledge in all cases. It is for this reason that when analyzing the simulated data, we also consider a threshold of the 90^{th }percentile of the data allowing for an ad hoc databased determination of a threshold.
Tests of H_{0,A }(WangAllison tests)
For comparative purposes, the first category of tests we evaluated were the tests denoted QT3 and QT4 in Wang et al [5] which are, respectively, Boschloo's test and an exact unconditional test based on the observed difference divided by its estimated standard error under the null hypothesis (score statistic) and are described in more detail by Mehrotra et al. [7]. These were the two tests that Wang et al. [5] had found performed best as tests of H_{0,A}.
Tests of H_{0,B}
In testing H_{0,B}, subjects were only included in the analysis when their lifespans exceeded τ. Distributions of survival times (lifespans) are rarely Gaussian and, even if they were nearly Gaussian after, for example, log transformation, the distribution of just the tail portion (i.e., f (YY > τ) would not be. Hence, in constructing tests we relied on nonparametric statistical methods. Specifically, we used the WilcoxonMannWhitney (exact) test [11,12] and a permutation test (with tstatistic) as described by Good [13] to test for differences in lifespan among those subjects whose lifespans exceeded τ.
Tests of H_{0,C }(new tests)
In testing H_{0,C}, all subjects were analyzed, but the variable analyzed was Z as defined above. Because the distribution of Z cannot be normally distributed, we again used the WilcoxonMannWhitney test and a permutation test to test for differences in Z.
For a dataset with n_{1 }(n_{2}) subjects in treatment (control) group, the permutation test can be performed in the following way: First put all the (n_{1 }+n_{2}) subjects together, and then generate 1000 replicated datasets. For each replicated dataset, we randomly sample n_{1 }subjects from the (n_{1 }+n_{2}) subjects and assign them to treatment group, and assign the left n_{2 }subjects to control group. We run Ttest on the observed dataset and the 1000 replicated datasets. Let T_{0 }be the T value for the observed dataset, then pvalue for the permutation test is calculated as the proportion of replicated datasets with absolute T values greater than or equal to the absolute valued of T_{0}.
Results
Results are displayed in Tables 1 to 5. As can be seen, the new methods for tests of H_{0,C }controls type I error rates quite well. The power of the new methods are always higher than or very close to that of the methods for tests of H_{0,A }(WangAllison tests) and are higher than that of the methods for tests of H_{0,B }(WilcoxonMannWhitney tests and permutation tests for observations above the threshold τ) in some of the simulations.
Table 1. Performance (type 1 error rates) of the tests in simulation 1 under H_{0,C }(i.e., both H_{0,A }and H_{0,B }are true) and yet f (YY ≤ τ ∩ X = 1) is radically different from f (YY ≤ τ ∩ X = 0) (see Figure 3 for details of simulation).
Table 2. Performance of the tests in simulation 2, H_{0,A }is true, H_{0,B }is false and f (YY ≤ τ ∩ X = 1) is radically different from f (YY ≤ τ ∩ X = 0) (see Figure 3 for details of simulation).
Table 3. Performance of the tests in simulation 3, H_{0,B }is true, H_{0,A }is false and f (YY ≤ τ ∩ X = 1) is radically different from f (YY ≤ τ ∩ X = 0) (see Figure 3 for details of simulation).
Table 1 shows the type I error rate of the tests (in simulation 1) when the null hypothesis H_{0,C }is true (i.e., both H_{0,A }and H_{0,B }are true) and yet f (YY ≤ τ ∩ X = 1) is radically differentfrom f (YY ≤ τ ∩ X = 0). The type I error rates of the new methods are comparable to those of the methods for tests of H_{0,A }and those of the methods for tests of H_{0,B }. It is note worthy that there is a slight but fairly consistent excess of type I errors when the sample 90^{th }percentile is used rather than a fixed cutoff point. This is because the sample 90^{th }percentile is a random variable and when it falls below its population level, the null hypothesis is no longer strictly true in our simulations. That is, the tests remain valid tests of differences in distributions above the actual value used but should not be strictly interpreted as tests of differences in distributions above the 90^{th }(or any other percentile). In practice, this distinction is probably trivial.
In simulation 2 (see Table 2), where H_{0,A }is true, H_{0,B }is false and f (YY ≤ τ ∩ X = 1) is radically different from f (YY ≤ τ ∩ X = 0), the new methods for tests of H_{0,C }and the methods for tests of H_{0,A }have lower power than that of the corresponding methods for tests of H_{0,B}, however, the new methods for tests of H_{0,C }can slightly improve the power compared to the methods for tests of H_{0,A}.
Table 3 shows the power of the tests in Simulation 3, where H_{0,B }is true, H_{0,A }is false and f (YY ≤ τ ∩ X = 1) is radically different from f (YY ≤ τ ∩ X = 0). The new methods for tests of H_{0,C }and the methods for tests of H_{0,A }have very similar power which is much higher than that of the corresponding methods for tests of H_{0,B}.
From simulation 4 (see Table 4), where H_{0,B }is false, H_{0,A }is false and f (YY ≤ τ ∩ X = 1) and f (YY ≤ τ ∩ X = 0) are identical, we can find that the new methods for tests of H_{0,C }always have higher power than the corresponding methods for tests of H_{0,A}. When τ being set to the 90th percentile of the sample, the new methods also have higher power than the corresponding methods for tests of H_{0,B}.
Finally, we conducted a set of simulations under what we perceived to be the most realistic situations. Here both H_{0,A }and H_{0,B }are false, f (YY ≤ τ ∩ X = 1) is quite different from f (YY ≤ τ ∩ X = 0), and the distributions have no discontinuities. In other words, there is just a simple reduction in the hazard rate when X = 1. Table 5 presents the power of the tests in Simulation 5, where f (YX = 1) = 1.2f (YX = 0). In this simulation, the methods for tests of H_{0,B }almost have no power because the control group always has no or few observations above the threshold τ . The new methods for tests of H_{0,C}, when using a permutation test, have power higher than or equal to that of the methods for tests of H_{0,A}.
Illustration with real data
To illustrate the methods, we applied them to two real datasets. In both of these datasets, prior research had shown differences in overall survival rate and we tested for differences in 'maximum lifespan' herein. The first was a subset of data reported by Vasselli et al [10]. The subset of the data consists of two groups of SpragueDawley rats, those kept on a highfat diet ad libitum throughout life and becoming obese (EOHF) and those kept on a highfat diet ad libitum until earlymiddle adulthood, becoming obese, and subsequently reduced to normal weight via caloric restriction, but on the same highfat diet (WLHF). Each group had 49 rats (see Figure 4 for the histograms for the data). The second dataset was from a study comparing the lifespan of Agoutirelated proteindeficient (AgRP(/)) mice to wildtype mice (+/+) as reported by Redmann & Argyropoulos [14]. This dataset consists of 16 mice with genotype '+/+' and 21 mice with genotype '/' (see Figure 5 for the histograms for this dataset). From Figure 4, we can see the upper tails of the histograms of the two groups are different. Similar results can be found in Figure 5.
Results (p values of tests) are shown in Table 6. As can be seen, when setting τ equal to 110 (100) for the first (second) datasets, both the methods for tests of H_{0,A }and the new methods for tests of H_{0,C }can detect the differences in 'maximum lifespan' between groups at nominal alpha levels of 0.01 (0.05) for the first (second) datasets. But the methods for tests of H_{0,B }cannot detect the difference for all different values of τ . The following description may provide some explanation to these results. For the first dataset, when set τ = 110, the proportions of the observations greater than τ in the EOHF group and WLHF group (i.e., estimations of P(Y > τ  X = 0) and P(Y > τ  X = 1)) are 0.061 and 0.306, respectively. These two proportions are significantly different and not surprisingly, the methods for tests of H_{0,A }can detect the difference in 'maximum lifespan' between the two groups. Second, the sample means of the observations greater than τ in the two groups (i.e., estimations of μ (Y  Y > τ ∩ X = 1) and μ (Y  Y > τ ∩ X = 0)) are 117.8 and 122.9, respectively, and there is no much difference between these sample means. However the sample means of the Zvalues in the two group (i.e., the estimations of P(Z  X = 0) and P(Z  X = 1)) are 7.210 and 37.633, respectively, and are greatly different, where, Z_{i }≡ I(Y_{i }> τ)Y_{i}. These may explain that the methods for tests of H_{0,B }cannot reject the null but the new methods for tests of H_{0,C }can detect the difference in 'maximum lifespan' between the two groups. Similarly, for the second dataset, when set τ = 100, the proportions of the observations greater than τ in the group with genotype '+/+' and group with genotype '/' are 0.188 and 0.571, respectively. The sample means of the observations greater than τ in the two groups are 109.3 and 110.9, respectively. The sample means of the Zvalues in the two groups are 20.5 and 63.4 respectively.
Table 6. Results (p values of tests) of application to two real datasets.
From Table 6 we can also see that in almost all situations the pvalues of the new methods for tests of H_{0,C }are somewhat smaller than those of the methods for tests of H_{0,A}. This is consistent with the simulations showing greater power of the new methods.
Discussion
Herein, we proposed new methods for testing the difference of 'maximum' lifespan between groups (e.g., treatment and control). By defining a new variable Z such that Z_{i }≡ I (Y_{i }> τ)Y_{i }for each observation and then applying WilcoxonMannWhitney test or better still a permutation test to Z, the new methods achieve far better performance when considered across a broad range of circumstances in terms of both Type1 error rates and power. In the new methods, we use the WilcoxonMannWhitney test or permutation test. One could also choose to use a bootstrap test in place of these two tests. However, additional simulations would likely be warranted to evaluate its performance relative to the permutation test we have evaluated herein.
It is straightforward to extend the new methods to more than two groups. For example, one could use the KruskalWallis Test to replace the WilcoxonMannWhitney test, or use permutation testing for multiple groups to replace that for two groups.
We have shown that the new methods are effective by simulation studies when the sample size (N) of each group is 50, 100, or 200. We expect that these methods will be also be relatively more powerful than existing competitors for much larger sample sizes, such as N = 500 or even N = 5000. There are some mouse data sets (like those of the National Institute of Aging's Intervention Testing Program) where N > 500, and worm and fly data sets in which N may sometimes even exceed 5000. We expect that the new methods are equally applicable to the analysis of such data.
Finally, we note that the tests proposed here are described for the context of testing for differences in lifespan. However, there is nothing intrinsic to them that limits their use to survival data. They could be applied to any situation in which one wanted to test for group differences in the tails of distributions.
Competing interests
The authors declare they have no competing interests.
Authors' contributions
DBA participated in all parts of the work of the study (including the study design, methodology development, simulations, data acquisition, and manuscript drafting). He wrote major sections of the original manuscript. He revised final version of the manuscript. DTR provided consulting on the statistical issues in the study and manuscript editing. SZ provided assistance in programming for simulation studies. WW provided consulting on simulation and prepared the figures. GG did all simulation studies and real data analyses and drafted the sections of Results, Illustration with real data, and Discussion of the manuscript and participated in revision of the manuscript.
Acknowledgements
We thank Richard Miller, David Harrison, and Simon Klebanov for thought provoking dialogue that inspired this paper and George Argyropoulos for graciously providing data. This research was supported in part by NIH grants P30DK056336, R01DK067487, and P01AG11915 and by grant GM073766 from the National Institute of General Medical Sciences.
References

Miller RA, Harrison DE, Astle CM, Floyd RA, Flurkey K, Hensley KL, Javors MA, Leeuwenburgh C, Nelson JF, Ongini E, Nadon NL, Warner HR, Strong R: An aging Interventions Testing Program: study design and interim report.
Aging Cell 2007, 6(4):565575. PubMed Abstract  Publisher Full Text

Wanagat J, Allison DB, Weindruch R: Caloric intake and aging: mechanisms in rodents and a study in nonhuman primates.
Toxicol Sci 1999, 52(2 Suppl):3540. PubMed Abstract  Publisher Full Text

Ingram DK, Zhu M, Mamczarz J, Zou S, Lane MA, Roth GS, deCabo R: Calorie restriction mimetics: an emerging research field.
Aging Cell 2006, 5(2):97108. PubMed Abstract  Publisher Full Text

Weindruch R, Walford RL: The retardation of aging and disease by dietary restriction. In The retardation of aging and disease by dietary restriction. C.C. Thomas Publisher, Springfield, IL; 1988. PubMed Abstract

Wang C, Li Q, Redden DT, Weindruch R, Allison DB: Statistical methods for testing effects on "maximum lifespan".
Mech Ageing Dev 2004, 125(9):62932.
Erratum in: Mech Ageing Dev 2006, 127(7):652
PubMed Abstract  Publisher Full Text 
Redden DT, Fernandez JR, Allison DB: A simple significance test for quantile regression.
Stat Med 2004, 23(16):258797. PubMed Abstract  Publisher Full Text

Mehrotra DV, Chan IS, Berger RL: A cautionary note on exact unconditional inference for a difference between two independent binomial proportions.
Biometrics 2003, 59:441450. PubMed Abstract  Publisher Full Text

Little RC, Folks JL: On the comparison of two methods of combining independent tests.
Journal of the American Statistical Association 1972, 67(337):223. Publisher Full Text

Dominici F, Zeger SL: Smooth quantile ratio estimation with regression: estimating medical expenditures for smokingattributable diseases.
Biostatistics 2005, 6:505519. PubMed Abstract  Publisher Full Text

Vasselli JR, Weindruch R, Heymsfield SB, PiSunyer FX, Boozer CN, Yi N, Wang C, Pietrobelli A, Allison DB: Intentional weight loss reduces mortality rate in a rodent model of dietary obesity.
Obes Res 2005, 13(4):693702. PubMed Abstract  Publisher Full Text

Wilcoxon F: Individual comparisons by ranking methods.
Biometrics 1945, 1:8083. Publisher Full Text

Mann HB, Whitney DR: On a test of whether one of two random variables is stochastically larger than the other.
Annals of Mathematical Statistics 1947, 18:5060. Publisher Full Text

Good P: Permutation Tests: A Practical Guide to Resampling Methods for Testing Hypotheses. New York: SpringerVerlag; 1994.

Redmann SM Jr, Argyropoulos G: AgRPdeficiency could lead to increased lifespan.
Biochem Biophys Res Commun 2006, 351(4):8604. PubMed Abstract  Publisher Full Text  PubMed Central Full Text
Prepublication history
The prepublication history for this paper can be accessed here: