Palynological results from Julietta Lake currently provide the most direct evidence to support the existence of a glacial refugium for Pinus pumila in mountains of southwestern Beringia. Both percentages and accumulation rates indicate the evergreen shrub survived until at least ∼ 19,000 14C yr BP in the Upper Kolyma region. Percentage data suggest numbers dwindled into the late glaciation, whereas pollen accumulation rates point towards a more rapid demise shortly after ∼ 19,000 14C yr BP. Pinus pumila did not re-establish in any great numbers until ∼ 8100 14C yr BP, despite the local presence ∼ 9800 14C yr BP of Larixdahurica, which shares similar summer temperature requirements. The postglacial thermal maximum (in Beringia ∼ 11,000-9000 14C yr BP) provided Pinus pumila shrubs with equally harsh albeit different conditions for survival than those present during the LGM. Regional records indicate that in this time of maximum warmth Pinus pumila likely sheltered in a second, lower-elevation refugium. Paleoclimatic models and modern ecology suggest that shifts in the nature of seasonal transitions and not only seasonal extremes have played important roles in the history of Pinus pumila over the last ∼ 21,000 14C yr BP.

Background. Little is known about the extent to which negative life events predict depressive symptoms in ethnically diverse groups or whether this relationship is proximal or enduring.

Method. The relationship between negative life events in adolescence and depressive symptoms in young adulthood was studied in a sample of over 1300 black and white female adolescents. Five domains of life events were assessed at age 16 years and depressive symptoms were measured at age 18 and again at age 21 years. Questions of interest included whether the association continued over time and whether there were specific domains of life events that predicted symptoms better than others.

Results. The total number of negative life events at time 1 predicted depressive symptoms at both time 2 and time 3. Interpersonal loss events and other adversities, however, predicted depressive symptoms only at time 2, whereas at time 3, only interpersonal trauma was a significant predictor. No ethnic differences were found, indicating that the relationship between life events and depressive symptoms appears to be similar for black and white adolescent girls.

Conclusions. The results suggest that negative life events and some specific type of stressors increase the likelihood of the onset of depression symptoms in future years, for both black and white girls. Early preventive efforts should be directed at adolescents who experience loss due to death of a significant other, traumatic events, and psychosocial adversities to forestall the development of depressive symptoms.

This paper describes a method to construct a standardized health care resource use database. Billing and clinical data were analyzed for 916 patients who received liver transplantations at three medical centers over a 4-year period. Data were checked for completeness by assessing whether each patient's bill included charges covering specified dates and for specific services, and for accuracy by comparing a sample of bills to medical records. Detailed services were matched to a standardized service list from one of the centers, and a single price list was applied. For certain services, clinical data were used to estimate service use or, if a match was not possible, adjusted charges for the services were used. Twenty-three patients were eliminated from the database because of incomplete resource use data. There was very good correspondence between bills and medical records, except for blood products. Direct matches to the standardized service list accounted for 69.3% of services overall; 9.4% of services could not be matched to the standardized service list and were thus adjusted for center and/or time period. Clinical data were used to estimate resource use for blood products, operating room time, and medications; these estimations accounted for 21.3% of services overall.

A database can be constructed that allows comparison of standardized resource use and avoids biases due to accounting, geographic, or temporal factors. Clinical data are essential for the creation of such a database. The methods described are particularly useful in studies of the cost-effectiveness of medical technologies.

The present study set out to examine the relationship between marital status, poverty and depression in a sample of inner-city women. Single and married mothers were followed up over a 2-year period during which time rates of psychosocial risk factors, onset of depression and experience of chronic episodes were measured. Risk of onset was double among single mothers. Single mothers were twice as likely as their married counterparts to be in financial hardship, despite being twice as likely to be in full-time employment. Both of these factors were independently associated with onset in single mothers. The link between them and onset was via their association with humiliating or entrapping severe life events. Single parents were at a much raised risk of experiencing these events. Onset was also more likely to follow such an event when women had poor self-esteem and lack of support, both of which were more common among single mothers. These risk factors were more frequently found among those in financial hardship. Financial hardship was also related to risk of having a chronic episode (lasting at least a year), of which single parents were also at greater risk. The majority of chronic episodes among single mothers had their origins in prior marital difficulties or widowhood and rates of chronicity reduced with length of time spent in single parenthood. Results are discussed in terms of an aetiological model of onset in which financial hardship probably influences outcome at a wide variety of points.

In ten lambs (average live weight 33 kg), five offered 300 g/d (approximately 0.6 x maintenance; L) and five 900 g/d (1.8 x maintenance; H), tissue protein synthesis was measured by three procedures simultaneously. The techniques involved continuous infusion of [U-14C]phenylalanine and [1-13C]leucine over 7–8 h followed by a terminal large dose of [15N]phenylalanine during the last 30 or 60 min. Rates of protein synthesis were then calculated based on the free amino acid or oxo-acid isotopic activity in either arterial, iliac venous blood or tissue homogenate for the continuous-infusion studies, or on plasma or tissue homogenate for the large-dose procedure. For muscle (> 99%), and to a lesser extent skin (85–93%), effective flood conditions were achieved with the [15N]phenylalanine but were either not established or maintained for liver and tissues of the gastrointestinal tract (< 50%). The large dose of phenylalanine also caused changes in the concentration and isotopic activity of blood leucine and 4-methyl-2-oxo-pentanoate. Based on the assumption that the large-dose procedure yields the closest value for the true rate of protein synthesis (L 1.97%/d, H 2.85%/d) then, for muscle, only values based on the homogenate as precursor gave comparable results for both leucine (L 1.83%/d, H 3.01%/d) and phenylalanine (L 1.67%/d, H 2.71%/d) continuous infusion. The values based on the arterial or venous amino or oxo-acid were significantly less, more so at the lower intake. In contrast, for skin, a tissue dominated by export protein synthesis, values from the large-dose procedure (L 6.37%/d, H 10.98%/d) were similar to those derived with arterial or venous metabolites as precursor (L 5.23 and 6.93%/d, H 9.98 and 11.71%/d for leucine), but much less than those based on homogenate data. Based on the large-dose technique, protein synthesis increased with intake in muscle (P < 0.001), skin (P = 0.009) and liver (26.7 v. 30.5%/d; P = 0.029). The contributions of muscle and skin to total protein synthesis were approximately equal. The incremental efficiency of conversion for muscle of synthesized protein into deposition appeared to be similar to values reported for rodents