A subgroup analysis of treatment efficacy according to whether patients had low-, intermediate-, or high-risk disease, showed that the addition of short-term ADT to radiotherapy conferred the greatest clinical benefit in the intermediate-risk subgroup, with an increase in the 10-year rate of overall survival from 54 to 61% (hazard ratio for death with radiotherapy alone, 1.23; 95% CI, 1.02 to 1.49) and a reduction in the 10-year disease-specific mortality from 10 to 3% (hazard ratio, 2.49; 95% CI, 1.50 to 4.11). No significant benefit was shown in the low-risk subgroup, with an increase in the 10-year rate of overall survival from 64 to 67% (hazard ratio for death with radiotherapy alone, 1.07; 95% CI, 0.83 to 1.39) and an increase in the 10-year disease-specific mortality from 1 to 3% (hazard ratio, 0.63; 95% CI, 0.21 to 1.92). An interaction test revealed no significant interaction effect between treatment and risk category for overall survival (P = 0.71) and only a weak suggestion of a differential benefit according to risk group for disease-specific mortality (P = 0.08). In all three risk subgroups, short-term ADT was associated with a significant reduction in biochemical failure (Table 2). The incidence of positive findings on repeat prostate biopsy in the lowrisk, intermediate-risk, and high-risk subgroups was fairly uniform in the radiotherapy-alone group, at 35%, 41%, and 39%, respectively, as compared with 12%, 24%, and 30% in the combined-therapy group. Analyses of treatment efficacy separately in white and black patients and in patients who were 70 years of age or younger and those who were older than 70 years were also performed.

    The addition of short-term ADT was associated with a benefit in all these subgroups, with the 10-year rate of overall survival increasing from 57 to 62% among white patients (hazard ratio for death with radiotherapy alone, 1.19), 55 to 61% among black patients (hazard ratio, 1.15), 64 to 70% among patients who were 70 years of age or younger (hazard ratio, 1.23), and 50 to 54% among those older than 70 years of age (hazard ratio, 1.11), with no statistical evidence of a differential benefit between whites and blacks (interaction test, P = 0.79) or between age subgroups (P = 0.47). Among black patients, the addition of shortterm ADT to radiotherapy was associated with a decrease in the 10-year disease-specific mortality from 7 to 5% (hazard ratio with radiotherapy alone, 1.27) and a decrease in the 10-year rate of biochemical failure from 40 to 19% (hazard ratio, 2.27).

    The multivariate analysis showed that a Gleason score of 7 or higher was a negative prognostic factor for overall survival, disease-specific mortality, distant metastases, and biochemical failure. Other identified negative prognostic factors were older age and nonwhite race or ethnic group for overall survival, clinical T2 lesions for diseasespecific mortality, and a PSA level of 4 ng per milliliter or higher for biochemical failure. A total of 439 patients (44%) in the combinedtherapy group and 404 (41%) in the radiotherapyalone group underwent a repeat prostate biopsy at 2 years. The initial Gleason scores, PSA values, and rates of biochemical failure at 2 years were similar between the patients who underwent biopsy and those who did not undergo biopsy. Persistent cancer was detected in 20% of the biopsy specimens in the combined-therapy group as compared with 39% in the radiotherapy-alone group (P<0.001).

    Patient-Reported Erectile Dysfunction
    At the pretreatment, 1-year, and 2-year evaluations, the Sexual Adjustment Questionnaire completion rates were 88%, 70%, and 27%, respectively. Before treatment, 48% of the respondents in the combined-therapy group and 54% of those in the radiotherapy-alone group reported that they were “always or almost always able to have an erection” (P = 0.15); the respective rates at 1 year were 21% and 31% (P = 0.004) (Table 3). Scores at 1 year, as compared with the pretreatment scores, were improved in 9% of the patients, the same in 33%, and worse in 58%, with no significant differences between the groups.

    Toxic Effects
    In the group treated with short-term ADT, the proportions of patients who had acute hepatic toxic effects (occurring up to 90 days after the start of radiotherapy) of grade 1, 2, 3, and 4 were 20%, 5%, 3%, and less than 1%, respectively; late hepatic toxic effects were seen in 4%, 1%, less than 1%, and 0 of these patients, respectively, as compared with 1%, 0, 0, and 0 in the radiotherapy-alone group. In both groups, the incidences of grade 3 or higher acute and late gastrointestinal toxic effects were 1% and 3%, respectively, with grade 5 toxic effects in three patients; two patients receiving radiotherapy alone died of obstruction of the colon, and one patient treated with radiotherapy plus short-term ADT died of colorectal bleeding. Acute grade 3 or higher genitourinary toxic effects were seen in 2% of patients in both groups, with late toxic effects in 8% of patients in the combined-therapy group and 6% of those in the radiotherapy-alone group. During the 8 weeks of short-term ADT before the start of radiotherapy (in the combined-therapy group), 55% of patients had hot flashes, 3% had rash, and the incidences of hepatic toxic effects, decreased hemoglobin levels, and elevated whitecell counts were 16%, 16%, and 4%, respectively (all grade 1). Grade 1 cardiac toxic effects were observed in 11 patients (1%) within 2 years after treatment.

    Statistical Analysis
    On the basis of previous studies, we expected patients treated with radiotherapy alone to have an 8-year overall survival rate of 60%. Adding short-term ADT was projected to increase this rate to at least 67%. Accordingly, the trial was designed to provide 90% power to detect a 7-percentage-point absolute difference in the 8-year survival rate, with the use of a one-sided log-rank test at the 0.025 significance level, requiring 1980 patients and 716 deaths for definitive analysis. We conducted three planned interim analyses with a significance level of P<0.001 as the criterion for early stopping, which was not met in any of these analyses. The primary end point, overall survival, was estimated by means of the Kaplan–Meier approach, and in the multivariate analyses, hazard ratios with 95% confidence intervals were estimated
    with the use of the Cox regression model.

    The end points of disease-specific mortality, distant metastases, and biochemical failure were estimated by means of the cumulative incidence function to account for competing risks. The Fine–Gray model was used to estimate hazard ratios for competing risks. The chi-square test was used to test differences in patients’ responses to the Sexual Adjustment Questionnaire. Three subgroup analyses of treatment efficacy were conducted. One was planned: a comparison of treatments within racial groups (white and black). Two were unplanned; one compared treatments within three risk categories defined according to baseline characteristics, and the other evaluated treatments within two age groups (≤70 years and >70 years). The likelihood-ratio test was used to assess whether there was a statistically significant difference in the magnitude of treatment benefit (i.e., interaction effect) according to patient subgroups.

    Characteristics of the Patients
    Between October 1994 and April 2001, a total of 2028 patients from 212 centers in the United States and Canada were randomly assigned to radiotherapy plus short-term ADT (the combined-therapy group) or radiotherapy alone. Forty-nine patients were ineligible, withdrew consent, or were lacking pretreatment data, leaving 1979 eligible patients who were available for evaluation (992 in the radiotherapy-alone group and 987 in the combined- therapy group).

    Compliance
    Compliance with the radiotherapy protocol was assessed in a random sample of 61% of the patients in the combined-therapy group and 64% in the radiotherapy-alone group. Compliance was balanced between the two treatment groups; 65% of the patients were treated per protocol, 19% were treated with acceptable variations, and 5% were treated with unacceptable variations. Data were incomplete in 1% of the patients because of death or progressive disease or because the patient declined radiotherapy, and 9% were not available for evaluation. Compliance with hormonal therapy was reviewed in all randomly assigned patients; the therapy was delivered per protocol in 78% of the patients, with acceptable variation in 17% and unacceptable deviation in 4%. Data were incomplete or were not available for evaluation in 1% of these patients.

    Outcomes
    The median follow-up for surviving patients was 9.1 years (range, 0.01 to 13.5) in the group of patients who received radiotherapy plus short-term ADT and 9.2 years (range, 0.2 to 14.1) in the group of patients who received radiotherapy alone. The 10-year rate of overall survival was 57% in the radiotherapy-alone group and 62% in the combined-therapy group (hazard ratio for death with radiotherapy alone, 1.17; 95% confidence interval [CI], 1.01 to 1.35; P = 0.03). The 10-year disease-specific mortality was 8% in the radiotherapy-alone group and 4% in the combined-therapy group (hazard ratio, 1.87; 95% CI, 1.27 to 2.74; P = 0.001). The 10-year rate of biochemical failure was 41% in the radiotherapy-alone group and 26% in the combined-therapy group (hazard ratio, 1.74; 95% CI, 1.48 to 2.04; P<0.001). The 10-year cumulative incidence of distant metastases was 8% in the radiotherapy-alone group and 6% in the combined-therapy group (hazard ratio, 1.45; 95% CI, 1.03 to 2.06; P = 0.04). The 10-year cumulative incidence of death from causes other than prostate cancer was 37% in the radiotherapy-alone group and 34% in the combined-therapy group (P = 0.56).

    Treatment
    All patients began treatment within 21 days after randomization. Radiotherapy, administered in daily 1.8-Gy fractions prescribed to the isocenter of the treatment volume, consisted of 46.8 Gy delivered to the pelvis (prostate and regional lymph nodes), followed by 19.8 Gy to the prostate, for a total dose of 66.6 Gy. Treatment of the regional lymph nodes was omitted in patients with negative lymph-node dissections or with a PSA level of less than 10 ng per milliliter and a Gleason score of less than 6. The study cochairs reviewed the simulation and portal films for each treatment field. Patients assigned to short-term ADT received flutamide at a dose of 250 mg orally three times a day and either monthly subcutaneous goserelin at a dose of 3.6 mg or intramuscular leuprolide at a dose of 7.5 mg for 4 months. Radiotherapy commenced after 2 months of androgen deprivation. Flutamide was discontinued if the level of alanine aminotransferase increased to more than twice the upper limit of the normal range.

    Assessments
    At the beginning and end of radiotherapy, assessments included a history taking and physical examination, performance status, complete blood count, and levels of alkaline phosphatase, alanine aminotransferase, PSA, and serum testosterone. Follow-up visits occurred at intervals of 3 months during the first year, 4 months during the second year, 6 months in years 3 through 5, and then annually. PSA values were obtained at each visit, along with the serum testosterone level and complete blood count during the first 2 years and the alkaline phosphatase level yearly. Repeat prostate biopsy 2 years after treatment was planned for patients without medical contraindications or evidence of local or distant disease and for patients who had not undergone orchiectomy or received hormonal treatment. Acute and late toxic effects were assessed with the use of the RTOG toxicity scales.

    At each visit during the first 2 years, the first 793 patients enrolled in the study completed the Sexual Adjustment Questionnaire. Erectile dysfunction was assessed with the question, “When sexually excited, are you able to get an erection?” The five levels of response were: always or almost always, sometimes, almost never or never, did not try, and no answer.

    End Points
    All end points were measured from the date of randomization. Overall survival, the primary end point, was calculated at the date of death from any cause. Secondary end points included diseasespecific mortality, distant metastases, biochemical failure (an increasing level of PSA), and the rate of positive findings on repeat prostate biopsy at 2 years. Disease-specific mortality included all deaths from prostate cancer or treatment complications, as well as deaths from unknown causes in patients with either active cancer or a previously documented relapse. The study cochairs reviewed the reported causes of death, and complicated cases were reviewed by committee. The scoring of distant metastasis required documentation of metastatic disease. The Phoenix Consensus Conference definition (an increase in the PSA level of >2 ng per milliliter above the nadir) was used to define biochemical failure.

    The introduction of prostate-specific antigen (PSA) testing has resulted in increased diagnoses of early-stage disease. Less is known about the role of short-term ADT in men receiving radiotherapy for these cancers. Accordingly, in 1994, the RTOG opened a large, randomized trial, RTOG 94-08, to evaluate whether adding short-term ADT to radiotherapy would improve survival among patients with nonbulky localized prostate adenocarcinomas and an initial PSA level of 20 ng per milliliter or less.

    Patients
    Patients with histologically confirmed prostate adenocarcinoma, stage T1b, T1c, T2a, or T2b (according to the 1992 classification of the American Joint Committee on Cancer), and a PSA level of 20 ng per milliliter or less were eligible for this international phase 3 study. Pretreatment evaluation included a digital rectal examination and bone scan. The regional lymph nodes were evaluated surgically by means of lymph-node sampling or clinically by means of lymphangiography or pelvic computed tomography. The Gleason score (the sum of the two most common histologic patterns or grades in a prostate tumor, each of which is graded on a scale of 1 to 5, with 5 indicating the most aggressive pattern) was determined, and tumors were also classified as well differentiated, moderately differentiated, or poorly differentiated. Eligibility criteria included a Karnofsky performance score of 70 or more (on a scale of 0 to 100, with higher scores indicating better performance status), an alanine aminotransferase level that was no more than twice the upper limit of the normal range, no evidence of regional lymph-node involvement or distant metastatic disease, and no previous chemotherapy, radiotherapy, hormonal therapy, cryosurgery, or definitive surgery for prostate cancer. Patients with previous basal-cell or squamous- cell skin carcinomas who had been diseasefree for 2 years or more before study entry, and patients with invasive cancers who had been disease- free for 5 years or more, were eligible if their participation was approved by the study cochairs. The institutional review boards of the participating institutions approved the study protocol, and all patients provided written informed consent. The National Cancer Institute sponsored the study. The drugs were purchased from vendors. No commercial support was provided for this study.

    Study Design
    After stratification according to PSA level (<4 vs. 4 to 20 ng per milliliter), tumor grade (well differentiated, moderately differentiated, or poorly differentiated), and surgical versus clinical documentation of negative regional nodal status, patients were randomly assigned to receive either radiotherapy plus short-term ADT or radiotherapy alone, according to the permuted-block randomization method described by Zelen. The RTOG carried out this trial and was responsible for data collection, statistical analysis, study design, and preparation of the manuscript.

    BACKGROUND
    It is not known whether short-term androgen-deprivation therapy (ADT) before and during radiotherapy improves cancer control and overall survival among patients with early, localized prostate adenocarcinoma.

    METHODS
    From 1994 through 2001, we randomly assigned 1979 eligible patients with stage T1b, T1c, T2a, or T2b prostate adenocarcinoma and a prostate-specific antigen (PSA) level of 20 ng per milliliter or less to radiotherapy alone (992 patients) or radiotherapy with 4 months of total androgen suppression starting 2 months before radiotherapy (radiotherapy plus short-term ADT, 987 patients). The primary end point was overall survival. Secondary end points included disease-specific mortality, distant metastases, biochemical failure (an increasing level of PSA), and the rate of positive findings on repeat prostate biopsy at 2 years.

    RESULTS
    The median follow-up period was 9.1 years. The 10-year rate of overall survival was 62% among patients receiving radiotherapy plus short-term ADT (the combined-therapy group), as compared with 57% among patients receiving radiotherapy alone (hazard ratio for death with radiotherapy alone, 1.17; P = 0.03). The addition of short-term ADT was associated with a decrease in the 10-year disease-specific mortality from 8% to 4% (hazard ratio for radiotherapy alone, 1.87; P = 0.001). Biochemical failure, distant metastases, and the rate of positive findings on repeat prostate biopsy at 2 years were significantly improved with radiotherapy plus short-term ADT. Acute and late radiationinduced toxic effects were similar in the two groups. The incidence of grade 3 or higher hormone-related toxic effects was less than 5%. Reanalysis according to risk showed reductions in overall and disease-specific mortality primarily among intermediate- risk patients, with no significant reductions among low-risk patients.

    CONCLUSIONS
    Among patients with stage T1b, T1c, T2a, or T2b prostate adenocarcinoma and a PSA level of 20 ng per milliliter or less, the use of short-term ADT for 4 months before and during radiotherapy was associated with significantly decreased disease-specific mortality and increased overall survival. According to post hoc risk analysis, the benefit was mainly seen in intermediate-risk, but not low-risk, men.

    In the 1980s, advances in both surgery and radiotherapy for clinically localized prostate cancer led to their acceptance as successful treatments, with considerable reductions in harmful side effects as compared with earlier treatments. In the 1990s, reversible androgen suppression with the use of luteinizing hormone–releasing hormone analogues and oral antiandrogen agents was shown to induce apoptotic regression in androgen- responsive cancers, potentially improving the prospects of local control and the duration of survival free of metastatic disease. Among patients with locally advanced disease, phase 3 clinical trials showed that when added to radiotherapy, long-term treatment with these agents (≥2 years) improved overall survival but also increased toxic effects, including erectile dysfunction and myocardial infarction.5 Short-term androgendeprivation therapy (ADT) could potentially mitigate these toxic effects. A Radiation Therapy Oncology Group (RTOG) phase 3 clinical trial, reported in 1994, showed that short-term ADT administered for 4 months before and during radiation therapy significantly improved local control and disease-free survival among patients with bulky stage T2c to T4 tumors. Other trials have also shown benefits from this approach.

    National efforts are under way to link together large administrative databases to permit hypotheses concerning adverse associations to be tested more rigorously. To enable medical sleuths to detect the initial scent that leads them to track a lowfrequency adverse drug effect, more open and better reporting is needed. But fuller reporting of all adverse experiences without filtering on the basis of statistical significance or perceived causality would result in the publication of more tedious supplemental tables that primarily contribute to information overload. Similarly, Justice Sotomayor noted that such unfiltered information “could bury the shareholder in an avalanche of trivial information.”

    However, in cases in which data showing initially seemingly unimportant imbalances eventually add up to a clear signal of an adverse action of a drug, transparent early reports could reduce the likelihood of litigious arguments concerning who knew what and when. Individual manufacturers should not be in the position of determining what information is considered material for public dissemination. Physicians are also well aware that our noninfallible but important recommendations are based on our best assessments of incomplete data, with levels of evidence ranging from firm to anecdotal. With the finality afforded by the Supreme Court decision in Matrixx, investors can be assured of increased access to statistically nonsignificant information regarding reports of adverse drug experiences. Although the medical establishment lacks legal authority, it could use its standards- setting powers to improve access to the same level of information. The resulting flood of data, though likely to represent biologic noise rather than evidence sufficient to establish even “probable cause,” would contribute to the total mix of available information and might, under some circumstances, influence reasonable prescribers and patients to alter their treatment plans: a sommelier, for example, might consider any report of anosmia to be material.

    Dr. Pfeffer is a physician in the Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital, and a professor at Harvard Medical School; and Ms. Bowler is a U.S. magistrate judge, U.S. District Court — all in Boston.

    In the recent unanimous decision in Matrixx Initiatives v. Siracusano, the U.S. Supreme Court applied the “fair preponderance of the evidence” standard of proof used for civil matters, in which a particular conclusion is deemed “more likely than not” to be justified. At issue was whether Matrixx had violated federal securities laws by failing to disclose to shareholders sporadic reports of anosmia associated with the use of its Zicam nasal spray before the Food and Drug Administration (FDA) issued a warning about that association in 2009. The question before the Court was not whether the drug caused the loss of smell, but rather whether the company failed to provide material information to the investor plaintiffs that would have led a “reasonable shareholder” to alter his or her investment strategy. The initial trial court was persuaded by the company’s primary argument that the evidence suggesting that its product caused anosmia did not reach statistical significance and therefore should not have been considered material. In upholding the ruling of the appellate court, which had reversed the trial court’s decision, the Supreme Court ruled that whether or not it was considered statistically significant, the information about the seemingly infrequent occurrences of loss of smell after use of the product was indeed material to investors. Speaking for the undivided Court, Justice Sonia Sotomayor also acknowledged that the mere existence of reports of adverse events associated with a drug does not prove causality — but asserted that such a high level of proof did not have to be achieved. Similarly, under the Code of Federal Regulations for the FDA, warnings and precautions regarding the safety of drugs must be revised to include information on “a clinically significant hazard as soon as there is reasonable evidence of a causal association with a drug; a causal relationship need not have been definitely established.” There is no requirement for statistical significance.

    Clinicians are well aware that to be considered material, information regarding drug safety does not have to reach the same level of certainty that we demand for demonstrating efficacy. We understand that clinical trials that are designed to prove that a drug is effective use preplanned statistical analyses focused on a specific, carefully defined and adjudicated primary end point. Moreover, the number of subjects who will have to experience this targeted event for researchers to adequately test whether it occurs at the same rate as it does in a comparison group (the trial’s statistical power) is also established before the study begins. This same carefully constructed statistical framework is not, and understandably cannot be, used for evaluating unplanned and uncommon adverse events. When studying safety, we search for signals of imbalances and attempt to piece together multiple underpowered comparisons to obtain a better estimate of the risk. Sorting the wheat of true adverse drug effects from the chaff of biologic variability and chance associations is exceedingly difficult. A staggering and increasing number of reports are received by the FDA’s Adverse Event Reporting System (AERS) each year — more than half a million in 2009.

    The legal and medical systems both strive for truth while acknowledging that there are no absolutes. Both systems require evidence, which they categorize in a hierarchy of levels, on which to base decisions that can have major effects on the quality and even quantity of people’s lives. In law, the strictest standard of proof applies in criminal matters, in which the presumption of innocence requires that guilt be established “beyond a reasonable doubt” to attempt to rule out the possibility of convicting an innocent person — though of course the application of this level of proof carries the risk of occasional acquittal of a defendant who is actually guilty. A lower, but still relatively stringent, standard of proof, that of “clear and convincing” evidence, applies to certain discrete civil matters and criminal matters such as the setting of bail. A still lower standard requiring conclusions based on a “fair preponderance of the evidence” applies in the great majority of civil matters, and an even lower standard in which only “probable cause” must be established permits certain criminal proceedings to be initiated. These levels of legal proof have analogies in medicine. Clinical trials use alpha (significance) levels, confidence intervals, and statistical power to gauge levels of certainty. To reject the null hypothesis (that a result occurred merely by chance) and deem an intervention effective in a clinical trial, the level of proof analogous to law’s “beyond a reasonable doubt” standard would require an extremely stringent alpha level to permit researchers to claim a statistically significant effect, with the offsetting risk that a truly effective intervention would sometimes be deemed ineffective. Instead, most randomized clinical trials are designed to achieve a somewhat lower level of evidence that in legal jargon might be called “clear and convincing,” making conclusions drawn from it highly probable or reasonably certain.

    Although errors can be made by both the judicial system and the medical research system, the former provides the opportunity to appeal a court’s decision, and in the latter, reproducibility or independent confirmation of a result greatly enhances the reliability of findings. Unlike the categorical decisions of the courts, which immediately carry the weight of the law regardless of their popularity, the results of a clinical trial can have greater or lesser impact depending on their eventual degree of acceptance by the medical community. The data from clinical trials are generally initially disseminated in peerreviewed medical journals, at scientific meetings, or both. The ultimate influence of a study then depends on the interpretation of the importance of its results by national guideline-setting committees, as well as by more local physician groups at journal clubs, morning reports, and rounds. In such scholarly dissections of the trial data, statistical tests represent only one aspect of the intense scrutiny applied in assessing the quality and robustness of the findings.

    A normal circadian rhythm cortisol pattern is one in which there is a rise before waking (before 7-8 AM), and then a gradual decline throughout the rest of the day(26). Twelve patients had recognized circadian patterns of cortisol fluctuation, which will hereafter be referred to as Cortisol Pattern 1 (CP1). Of the twelve with CP1 whom were classified in the “normal” pattern, six had normal values at all four test time points, as well as normal total cortisol values or burden. Thus, of the 29 total charts reviewed, only 21% of the patients would be classified with normal cortisol values as well as normal patterns. The remaining six patients with a “normal” pattern of cortisol secretion (high waking and then decreasing over the rest of the day) had abnormal values at one or more time points.

    Seventeen of the 29 total patients fell into our classification of dysregulated circadian cortisol patterns.

    This consisted of patients with cortisol patterns that did not decrease in slope over the course of the day. Fourteen of these 17 patients had cortisol values out of normal range at one or more time points, or in total cortisol burden. The dysregulated patients’ cortisol plots fell into three distinct patterns, which we will hereafter call Cortisol Pattern 2 (CP2), Cortisol Pattern 3 (CP3), and Cortisol Pattern 4 (CP4).
    Viagra online
    Patients grouped into CP1 begin with a burst of cortisol between 7 and 8 AM (between 13-23 nM), and drop off slowly throughout the day. Normal values are considered 4-8 nM between 11 AM and noon, 4-8 nM between 4 and 5 PM, and 1-3 nM between 11 PM and midnight. It is notable that with the group of patients who presented with normal patterns, there was a large range in cortisol levels they excreted at any specific point. Those patients classified with CP2 tended to have abnormal circadian cortisol levels, higher in the late morning (11 AM to noon), compared to higher waking levels. Their levels then dropped off or stayed the same later in the day and at midnight. In contrast, patients grouped into CP3 had peaks between 7-8 AM and a more significant drop at noon. This created a second “mini” peak on the graph between 4-5 PM, though the levels are actually closer to normal. Patients assigned with CP4 fell into a distinct pattern of increasing between 11 PM and midnight. While a normal value falls between 1-3 nM for this time point, these patients averaged 15.4 nM (SD 8.3). One of the patients had extremely high cortisol levels at two time points, including time point 4 (between 11 PM and midnight). If the data is reanalyzed without this patient, the average cortisol value is still high with an average of 6.75 nM (SD 2.06).

    Typically, the highest cortisol value of the day occurs at the waking time point. Yet, of the patient charts reviewed, if levels were low at all, they were low at this time point. One patient with low total cortisol had consistently low cortisol levels, except the late night point, which is typically the lowest cortisol level. There were three patients who had consistently high cortisol levels, and two of these also had high total cortisol values.