The association of smoking with outcomes following breast cancer prognosis is not well understood... more The association of smoking with outcomes following breast cancer prognosis is not well understood. In a cohort study called Life After Cancer Epidemiology (LACE), 2,265 women diagnosed with breast cancer were followed for a median of 12 years. We used multivariable proportional-hazards models to determine whether smoking, assessed approximately two years post-diagnosis, was associated with risk of death among these women. We also undertook a systematic review of all cohort studies to date that have examined the association between smoking and breast cancer mortality. Compared with never smokers, women who were current smokers had a twofold higher rate of dying from breast cancer [hazard ratio (HR) = 2.01, 95 % confidence interval (CI) 1.27-3.18] and an approximately fourfold higher rate of dying from competing (nonbreast cancer) causes (HR = 3.84, 95 % CI 2.50-5.89). Among seven studies that met the inclusion criteria in the systematic review, three studies and our own reported significantly increased risk of breast cancer death with current smoking. We found little evidence of an association between former smoking and breast cancer mortality (HR = 1.24, 95 % CI 0.94-1.64). Consistent with findings from our prospective observational study, the systematic review of seven additional studies indicates positive association of current smoking with breast cancer mortality, but weak association with former smoking. Women who smoke following breast cancer diagnosis and treatment are at higher risk of death both from breast cancer and other causes.
It is becoming increasingly evident that early-life events and exposures have important consequen... more It is becoming increasingly evident that early-life events and exposures have important consequences for cancer development later in life. However, epidemiological studies of early-life factors and cancer development later in life have had significant methodological challenges such as the long latency period, the distinctiveness of each cancer, and large number of subjects that must be studied, all likely to increase costs. These traditional hurdles might be mitigated by leveraging several existing large-scale prospective studies in the United States (US) and globally, as well as birth databases and birth cohorts, in order to launch both association and mechanistic studies of early-life exposures and cancer development later in life. Dedicated research funding will be needed to advance this paradigm shift in cancer research, and it seems justified by its potential to produce transformative understanding of how cancer develops over the life-course. This in turn has the potential to t...
Recent studies have suggested a role for dietary fat in the etiology of breast cancer. The relati... more Recent studies have suggested a role for dietary fat in the etiology of breast cancer. The relation of serum cholesterol and other serum lipid measures to breast cancer incidence was investigated in a cohort of 95, 179 women who had multiphasic health checkups from 1964 through 1972. Through 1977, 1,035 new breast cancer cases occurred in over 752,000 person-years of follow-up. Age-adjusted incidence rates were 1.45, 1.37, 1.31, and 1.40/1,000 person-years from the lowest to the highest quartile of serum cholesterol level, respectively. Similarly, no statistically significant relation was detected between beta-lipoprotein or total lipids and breast cancer. The sample size was sufficiently large to have detected a relative risk of 1.4 or larger with a probability of 99.9% at the 0.05 level of significance. The expected relation od breast cancer to established risk factors was confirmed by univariate analysis, and serum cholesterol and breast cancer were not associated after simultane...
The authors prospectively studied the incidence of cancers of the colon and rectum in 106,203 men... more The authors prospectively studied the incidence of cancers of the colon and rectum in 106,203 men and women, both white and black, who supplied data at northern California Kaiser Permanente facilities about use of alcoholic beverages in 1978-1984. Analysis controlling for age, sex, race, body mass index, coffee use, total serum cholesterol, and education showed a positive association of alcohol use to both types of cancer, which was stronger for rectal cancer (trend test, p = 0.03) than for colon cancer (trend test, p = 0.11). When persons with a daily intake of three or more drinks were compared with abstainers, relative risk for rectal cancer was 3.17 (95% confidence interval (CI): 1.05-9.57) and relative risk for colon cancer was 1.71 (95% CI: 0.92-3.19). Women with a daily intake of three or more drinks had a relative risk for colon cancer of 2.56 (95% CI: 1.03-6.40) compared with 1.16 (95% CI: 0.46-2.90) for men. Among drinkers, preference for wine, beer, or hard liquor had no ...
We examined the association between alcohol and caffeine consumption and premenstrual syndrome (P... more We examined the association between alcohol and caffeine consumption and premenstrual syndrome (PMS). One hundred two women with PMS matched by age and race to an equal number of women without PMS were compared for differences in intake of alcoholic and caffeinated beverages, with differences determined using Student's t-tests and conditional logistic regression for matched pairs. Information on alcohol and caffeinated beverage consumption was obtained from three 24-hour dietary recall interviews conducted during the postmenstrual period and from three conducted during the premenstrual period. Results showed that no significant difference was observed in total caffeine intake or in the individual caffeinated beverages consumed during either the post-menstrual or the premenstrual period. For alcohol consumption, however, women with PMS drank 1.41 (95% confidence interval; range, 0.34-2.47) more servings per week during the postmenstrual period. Based on post-menstrual consumption...
The rate of occurrence of stones of the urinary tract was assessed in a large population served b... more The rate of occurrence of stones of the urinary tract was assessed in a large population served by the Northern California Kaiser Foundation Health Plan. The study involved three separate groups. First, data were obtained by questionnaire from approximately 175,000 adults who took a multiphasic health checkup in the period 1964-1972; of these generally well adult members, 26.2/1000 persons (32.0/1000 men and 21.0/1000 women) reported having ever been told by a physician that they had a urinary tract stone. Second, data were obtained from 139,000 persons served by the San Francisco outpatient facility in 1970-1972; 1.22/1000 per year (1.81/1000 men and 0.59/1000 women) had an initial diagnosis of a "new or recurrent" stone of the upper urinary tract. The third set of data was procured from the entire Northern California region in 1971-1975; 0.36/1000 (0.52/1000 men and 0.19/1000 women) were discharged from a hospital each year with a diagnosis of upper urinary tract stone. ...
In a case-control study of urinalysis screening in the prevention of death from bladder cancer, h... more In a case-control study of urinalysis screening in the prevention of death from bladder cancer, hematuria was present in a higher proportion of cases than controls as long as five or six years before the diagnostic evaluation that led to the diagnosis of bladder cancer. In a separate cohort study data base that permitted the follow-up of 1046 persons with a physician's diagnosis of hematuria, 11 cases of bladder cancer were diagnosed more than two (mean 7.4) years after the hematuria diagnosis (4.3 cases expected; age-sex standardized morbidity ratio, 2.5; 95% confidence interval, 1.3-4.5). Bladder cancer was ruled out initially by cystoscopy in 8 of the 11 cases. Although we cannot be certain that preexisting bladder cancer or bladder cancer risk factors did not cause the bleeding, we hypothesize that hematuria can be a predictor as well as a manifestation of bladder cancer, based on a tendency for bladder mucosa with premalignant changes to bleed. The implications for screenin...
Barbiturate exposure during childhood was assessed from medical records of 237 children with intr... more Barbiturate exposure during childhood was assessed from medical records of 237 children with intracranial and spinal cord tumors and 474 matched controls in a prepaid health plan. In utero exposure was also examined in a subset of 86 "cases" and 172 controls whose mothers were health plan members during pregnancy. No association of in utero exposure to barbiturates was found [odds ratio (O.R.) = 0.96, 95% confidence interval (C.I.) = 0.47, 1.94]. An association was noted for barbiturate use during childhood (O.R. = 1.80, 95% C.I. = 1.18, 2.74) but was reduced (O.R. = 1.41, 95% C.I. = 0.89, 2.21) when history of epilepsy was taken into account and was no longer significant. An apparent dose-response effect disappeared after adjustment for a history of epilepsy. Although barbiturate use for epilepsy due to preexisting brain tumors clearly explains some of the observed association, the small, residual risk prevents us from ruling out a possible carcinogenic effect of barbitur...
To evaluate screening urinalysis for its efficacy in preventing fatal bladder cancer. A case-cont... more To evaluate screening urinalysis for its efficacy in preventing fatal bladder cancer. A case-control study of 290 case subjects with fatal bladder cancer upsilon 290 age and sex matched control subjects, with ascertainment of urinalyses by medical record review. In the five years before the symptoms that led to the detection of bladder cancer, case subjects had significantly fewer routine (screening) urinalyses than controls (matched odds ratio for one or more urinalyses upsilon none = 0.60 [95% confidence interval (CI) 0.41 to 0.87], adjusted for cigarette smoking and occupational bladder cancer risk). However urinalyses for urinary symptoms and problems before those that led to bladder cancer diagnosis were more frequent in the cases; hence there was only a small statistically non-significant deficit in cases of urinalyses for all reasons combined (corresponding adjusted odds ratio 0.94, 95% CI 0.61 to 1.46). Since urinalyses for symptoms and problems could have led to the detecti...
To investigate associations between maternal pregnancy hyperglycemia, gestational diabetes mellit... more To investigate associations between maternal pregnancy hyperglycemia, gestational diabetes mellitus (GDM), and offspring adiposity. We evaluated these associations in a longitudinal study of 421 mother-daughter pairs at Kaiser Permanente Northern California. Maternal pregnancy glucose values were obtained from maternal medical records. Outcomes included three measures of girls' adiposity, measured annually: (1) ≥85th age-specific percentile for BMI; (2) percent body fat (%BF); and (3) waist-to-height ratio (WHR). Adjusting for maternal age at delivery, race/ethnicity, pregravid BMI, girl's age, and girl's age at onset of puberty, having a mother with GDM increased a girl's risk of having a BMI ≥85th percentile or having %BF or WHR in the highest quartile (Q4), compared with those in the lowest quintile of blood glucose (odds ratio [OR] 3.56 [95% CI 1.28-9.92]; OR 3.13 [95% CI 1.08-9.09]; and OR 2.80 [95% CI 1.00-7.84], respectively). There was a significant interacti...
It has long been accepted that modern reproductive patterns are likely contributors to breast can... more It has long been accepted that modern reproductive patterns are likely contributors to breast cancer susceptibility because of their influence on hormones such as estrogen and the importance of these hormones in breast cancer. We conducted a meta-analysis to assess whether this 'evolutionary mismatch hypothesis' can explain susceptibility to both estrogen receptor positive (ER-positive) and estrogen receptor negative (ER-negative) cancer. Our meta-analysis includes a total of 33 studies and examines parity, age of first birth and age of menarche broken down by estrogen receptor status. We found that modern reproductive patterns are more closely linked to ER-positive than ER-negative breast cancer. Thus, the evolutionary mismatch hypothesis for breast cancer can account for ER-positive breast cancer susceptibility but not ER-negative breast cancer.
The objective of this study was to use successful quality improvement initiatives in large multis... more The objective of this study was to use successful quality improvement initiatives in large multispecialty medical groups to identify the organizational factors that were the most important to improvement. The study analyzed the most successful quality improvement initiatives from those submitted by the 24 members of the Council of Accountable Physician Practices. Twelve initiatives from 8 groups were selected that met the study criteria for large improvement for large numbers of patients. An independent group used these initiatives to identify potentially important factors and then asked key local leaders to rate the importance of these factors on a scale of 1 to 4, importance rating (1-4 scale) for each of 18 identified factors. Eighteen factors were identified and 5 stood out as ranked a 4 (Very Important) for at least 80% of the initiatives: Communication, Use of Evidence-Based Medicine, Leadership, Measurement, and Reporting. Another 7 of the 18 factors were ranked a 4 for more ...
In this paper, we describe the attributes of a comprehensive approach to breast cancer screening ... more In this paper, we describe the attributes of a comprehensive approach to breast cancer screening possible in an integrated health care system. We define an integrated health care system as one in which comprehensive preventive and medical care is provided to a defined population, by a defined panel of providers, and in which this care can be tracked using automated electronic data systems. Guided by the Pathways Conceptual Framework, it is possible to identify and systematically address (through research and interventions) the multiple predisposing, enabling, and reinforcing factors at the individual and organizational level associated with each step along the screening process. This framework is helpful as both a planning and an evaluation tool, in identifying places in the screening and follow-up process that could benefit from concerted quality improvement efforts and in guiding an evaluation of those efforts. We describe examples from research and organizational programmatic eff...
In preparation for jointly publishing official government cancer statistics, the Centers for Dise... more In preparation for jointly publishing official government cancer statistics, the Centers for Disease Control and Prevention (CDC) and the National Cancer Institute (NCI) compared incidence rates from NCI's Surveillance Epidemiology and End Results (SEER) Program and CDC's National Program of Cancer Registries (NPCR). Data for 1999 covering 78% of the US population were obtained from SEER and selected NPCR registries that met high quality data criteria. Incidence rates (per 100,000 population) were age-adjusted to the 2000 US standard population, and 95% gamma confidence intervals were estimated. NPCR rates for all sites combined were higher than SEER rates (males: NPCR 553.6, SEER 538.7; females: NPCR 420.8, SEER 412.5), but rates for specific cancer sites varied by registry program. Rates for colon cancer (males: NPCR 47.0, SEER 42.7; females: NPCR 36.5, SEER 33.8) and tobacco-related cancers were higher in NPCR than SEER. In contrast, NPCR rates were lower than SEER rates ...
Journal of the National Cancer Institute. Monographs, 1995
Current knowledge is scarce on Latino dietary practices. This study compared the dietary practice... more Current knowledge is scarce on Latino dietary practices. This study compared the dietary practices, alcohol consumption, and smoking behavior of Latinos and non-Latino whites in two randomly selected samples. Telephone surveys of adults 35-74 years of age from the Kaiser Permanente Medical Care Program (Latinos = 844; non-Latino whites = 510) and from census tract-based areas (Latinos = 806; non-Latino whites = 436) were conducted in the San Francisco Bay Area. Latino ethnicity was a significant predictor of dietary and alcohol consumption practices in multivariate logistic regression models after adjustment for sex, education, age, employment, health insurance, martial status, county of residence, and self-perceived health status. Compared with non-Latino whites, Latinos were significantly less likely to report eating vegetables and more likely to eat rice, beans, and fried foods and to drink whole milk. Less acculturated Latinos were more likely to eat fruits, rice, beans, meat, a...
Journal of the National Cancer Institute. Monographs, 1995
Difficulties in identification of Latinos from secondary datasets have often been addressed by th... more Difficulties in identification of Latinos from secondary datasets have often been addressed by the use of a Spanish surname list. To evaluate the sensitivity and specificity of the 1980 U.S. Census Spanish surname list in correctly identifying Latino ethnicity, we conducted a cross-sectional survey of adult members of a prepaid health plan in the San Francisco Bay Area. We compared a randomly selected list of 1345 Spanish surnames and 717 non-Spanish surnames from health plan membership lists with their self-identified race and ethnicity obtained by telephone interviews. The sample was stratified according to surname group (Spanish or non-Spanish), sex, age, and county. Compared with self-identified ethnicity, a Spanish surname was 88.4% sensitive in identifying Latino men and 70.4% sensitive in identifying Latina women. The non-Spanish surname list was 94.1% sensitive and 95.0% specific in correctly identifying non-Latinos. Although the overall negative predictive value was 97.8%, ...
The relation of work-related psychologic stress to spontaneous abortion was examined in a prospec... more The relation of work-related psychologic stress to spontaneous abortion was examined in a prospective study of 3,953 pregnant, employed members of a prepaid health plan in California, recruited between 1990 and 1991. Information on occupation, psychologic stress-related factors at work, social support at work, physical exertion at work, life events, pregnancy worries, and potential confounders was obtained by a telephone interview. Psychologic job stress and social support at work were assessed using an abbreviated version of the instrument developed by Karasek and colleagues (Los Angeles: University of Southern California, 1986). Results from the multivariate model indicated that, overall, stressful work was not associated with an increased risk of spontaneous abortion. However, interactions were observed between stressful work and maternal age over 32 years (p = 0.04), cigarette smoking (p = 0.02), and primigravidity (p = 0.06). Relative to the odds ratio for stressful work in you...
A case-control study of the relationship of serum cholesterol to large bowel cancer incidence was... more A case-control study of the relationship of serum cholesterol to large bowel cancer incidence was performed for 245 members of the Kaiser Permanente Medical Care Program in northern California. Five controls were obtained for each case, matched for age, sex, race, and time of examination. The mean serum cholesterol levels of cases were not significantly different from those of controls for all colorectal cancers, right-sided colon cancers, and left-sided colon cancers. Analysis by quartiles of serum cholesterol level suggested neither a direct nor an inverse relationship between serum cholesterol and large bowel cancer, and there was no evidence of a threshold value below which the risk of cancer increased. The study did not support the findings from other studies of an inverse relationship between serum cholesterol and large bowel cancer.
The association of smoking with outcomes following breast cancer prognosis is not well understood... more The association of smoking with outcomes following breast cancer prognosis is not well understood. In a cohort study called Life After Cancer Epidemiology (LACE), 2,265 women diagnosed with breast cancer were followed for a median of 12 years. We used multivariable proportional-hazards models to determine whether smoking, assessed approximately two years post-diagnosis, was associated with risk of death among these women. We also undertook a systematic review of all cohort studies to date that have examined the association between smoking and breast cancer mortality. Compared with never smokers, women who were current smokers had a twofold higher rate of dying from breast cancer [hazard ratio (HR) = 2.01, 95 % confidence interval (CI) 1.27-3.18] and an approximately fourfold higher rate of dying from competing (nonbreast cancer) causes (HR = 3.84, 95 % CI 2.50-5.89). Among seven studies that met the inclusion criteria in the systematic review, three studies and our own reported significantly increased risk of breast cancer death with current smoking. We found little evidence of an association between former smoking and breast cancer mortality (HR = 1.24, 95 % CI 0.94-1.64). Consistent with findings from our prospective observational study, the systematic review of seven additional studies indicates positive association of current smoking with breast cancer mortality, but weak association with former smoking. Women who smoke following breast cancer diagnosis and treatment are at higher risk of death both from breast cancer and other causes.
It is becoming increasingly evident that early-life events and exposures have important consequen... more It is becoming increasingly evident that early-life events and exposures have important consequences for cancer development later in life. However, epidemiological studies of early-life factors and cancer development later in life have had significant methodological challenges such as the long latency period, the distinctiveness of each cancer, and large number of subjects that must be studied, all likely to increase costs. These traditional hurdles might be mitigated by leveraging several existing large-scale prospective studies in the United States (US) and globally, as well as birth databases and birth cohorts, in order to launch both association and mechanistic studies of early-life exposures and cancer development later in life. Dedicated research funding will be needed to advance this paradigm shift in cancer research, and it seems justified by its potential to produce transformative understanding of how cancer develops over the life-course. This in turn has the potential to t...
Recent studies have suggested a role for dietary fat in the etiology of breast cancer. The relati... more Recent studies have suggested a role for dietary fat in the etiology of breast cancer. The relation of serum cholesterol and other serum lipid measures to breast cancer incidence was investigated in a cohort of 95, 179 women who had multiphasic health checkups from 1964 through 1972. Through 1977, 1,035 new breast cancer cases occurred in over 752,000 person-years of follow-up. Age-adjusted incidence rates were 1.45, 1.37, 1.31, and 1.40/1,000 person-years from the lowest to the highest quartile of serum cholesterol level, respectively. Similarly, no statistically significant relation was detected between beta-lipoprotein or total lipids and breast cancer. The sample size was sufficiently large to have detected a relative risk of 1.4 or larger with a probability of 99.9% at the 0.05 level of significance. The expected relation od breast cancer to established risk factors was confirmed by univariate analysis, and serum cholesterol and breast cancer were not associated after simultane...
The authors prospectively studied the incidence of cancers of the colon and rectum in 106,203 men... more The authors prospectively studied the incidence of cancers of the colon and rectum in 106,203 men and women, both white and black, who supplied data at northern California Kaiser Permanente facilities about use of alcoholic beverages in 1978-1984. Analysis controlling for age, sex, race, body mass index, coffee use, total serum cholesterol, and education showed a positive association of alcohol use to both types of cancer, which was stronger for rectal cancer (trend test, p = 0.03) than for colon cancer (trend test, p = 0.11). When persons with a daily intake of three or more drinks were compared with abstainers, relative risk for rectal cancer was 3.17 (95% confidence interval (CI): 1.05-9.57) and relative risk for colon cancer was 1.71 (95% CI: 0.92-3.19). Women with a daily intake of three or more drinks had a relative risk for colon cancer of 2.56 (95% CI: 1.03-6.40) compared with 1.16 (95% CI: 0.46-2.90) for men. Among drinkers, preference for wine, beer, or hard liquor had no ...
We examined the association between alcohol and caffeine consumption and premenstrual syndrome (P... more We examined the association between alcohol and caffeine consumption and premenstrual syndrome (PMS). One hundred two women with PMS matched by age and race to an equal number of women without PMS were compared for differences in intake of alcoholic and caffeinated beverages, with differences determined using Student's t-tests and conditional logistic regression for matched pairs. Information on alcohol and caffeinated beverage consumption was obtained from three 24-hour dietary recall interviews conducted during the postmenstrual period and from three conducted during the premenstrual period. Results showed that no significant difference was observed in total caffeine intake or in the individual caffeinated beverages consumed during either the post-menstrual or the premenstrual period. For alcohol consumption, however, women with PMS drank 1.41 (95% confidence interval; range, 0.34-2.47) more servings per week during the postmenstrual period. Based on post-menstrual consumption...
The rate of occurrence of stones of the urinary tract was assessed in a large population served b... more The rate of occurrence of stones of the urinary tract was assessed in a large population served by the Northern California Kaiser Foundation Health Plan. The study involved three separate groups. First, data were obtained by questionnaire from approximately 175,000 adults who took a multiphasic health checkup in the period 1964-1972; of these generally well adult members, 26.2/1000 persons (32.0/1000 men and 21.0/1000 women) reported having ever been told by a physician that they had a urinary tract stone. Second, data were obtained from 139,000 persons served by the San Francisco outpatient facility in 1970-1972; 1.22/1000 per year (1.81/1000 men and 0.59/1000 women) had an initial diagnosis of a "new or recurrent" stone of the upper urinary tract. The third set of data was procured from the entire Northern California region in 1971-1975; 0.36/1000 (0.52/1000 men and 0.19/1000 women) were discharged from a hospital each year with a diagnosis of upper urinary tract stone. ...
In a case-control study of urinalysis screening in the prevention of death from bladder cancer, h... more In a case-control study of urinalysis screening in the prevention of death from bladder cancer, hematuria was present in a higher proportion of cases than controls as long as five or six years before the diagnostic evaluation that led to the diagnosis of bladder cancer. In a separate cohort study data base that permitted the follow-up of 1046 persons with a physician's diagnosis of hematuria, 11 cases of bladder cancer were diagnosed more than two (mean 7.4) years after the hematuria diagnosis (4.3 cases expected; age-sex standardized morbidity ratio, 2.5; 95% confidence interval, 1.3-4.5). Bladder cancer was ruled out initially by cystoscopy in 8 of the 11 cases. Although we cannot be certain that preexisting bladder cancer or bladder cancer risk factors did not cause the bleeding, we hypothesize that hematuria can be a predictor as well as a manifestation of bladder cancer, based on a tendency for bladder mucosa with premalignant changes to bleed. The implications for screenin...
Barbiturate exposure during childhood was assessed from medical records of 237 children with intr... more Barbiturate exposure during childhood was assessed from medical records of 237 children with intracranial and spinal cord tumors and 474 matched controls in a prepaid health plan. In utero exposure was also examined in a subset of 86 "cases" and 172 controls whose mothers were health plan members during pregnancy. No association of in utero exposure to barbiturates was found [odds ratio (O.R.) = 0.96, 95% confidence interval (C.I.) = 0.47, 1.94]. An association was noted for barbiturate use during childhood (O.R. = 1.80, 95% C.I. = 1.18, 2.74) but was reduced (O.R. = 1.41, 95% C.I. = 0.89, 2.21) when history of epilepsy was taken into account and was no longer significant. An apparent dose-response effect disappeared after adjustment for a history of epilepsy. Although barbiturate use for epilepsy due to preexisting brain tumors clearly explains some of the observed association, the small, residual risk prevents us from ruling out a possible carcinogenic effect of barbitur...
To evaluate screening urinalysis for its efficacy in preventing fatal bladder cancer. A case-cont... more To evaluate screening urinalysis for its efficacy in preventing fatal bladder cancer. A case-control study of 290 case subjects with fatal bladder cancer upsilon 290 age and sex matched control subjects, with ascertainment of urinalyses by medical record review. In the five years before the symptoms that led to the detection of bladder cancer, case subjects had significantly fewer routine (screening) urinalyses than controls (matched odds ratio for one or more urinalyses upsilon none = 0.60 [95% confidence interval (CI) 0.41 to 0.87], adjusted for cigarette smoking and occupational bladder cancer risk). However urinalyses for urinary symptoms and problems before those that led to bladder cancer diagnosis were more frequent in the cases; hence there was only a small statistically non-significant deficit in cases of urinalyses for all reasons combined (corresponding adjusted odds ratio 0.94, 95% CI 0.61 to 1.46). Since urinalyses for symptoms and problems could have led to the detecti...
To investigate associations between maternal pregnancy hyperglycemia, gestational diabetes mellit... more To investigate associations between maternal pregnancy hyperglycemia, gestational diabetes mellitus (GDM), and offspring adiposity. We evaluated these associations in a longitudinal study of 421 mother-daughter pairs at Kaiser Permanente Northern California. Maternal pregnancy glucose values were obtained from maternal medical records. Outcomes included three measures of girls' adiposity, measured annually: (1) ≥85th age-specific percentile for BMI; (2) percent body fat (%BF); and (3) waist-to-height ratio (WHR). Adjusting for maternal age at delivery, race/ethnicity, pregravid BMI, girl's age, and girl's age at onset of puberty, having a mother with GDM increased a girl's risk of having a BMI ≥85th percentile or having %BF or WHR in the highest quartile (Q4), compared with those in the lowest quintile of blood glucose (odds ratio [OR] 3.56 [95% CI 1.28-9.92]; OR 3.13 [95% CI 1.08-9.09]; and OR 2.80 [95% CI 1.00-7.84], respectively). There was a significant interacti...
It has long been accepted that modern reproductive patterns are likely contributors to breast can... more It has long been accepted that modern reproductive patterns are likely contributors to breast cancer susceptibility because of their influence on hormones such as estrogen and the importance of these hormones in breast cancer. We conducted a meta-analysis to assess whether this 'evolutionary mismatch hypothesis' can explain susceptibility to both estrogen receptor positive (ER-positive) and estrogen receptor negative (ER-negative) cancer. Our meta-analysis includes a total of 33 studies and examines parity, age of first birth and age of menarche broken down by estrogen receptor status. We found that modern reproductive patterns are more closely linked to ER-positive than ER-negative breast cancer. Thus, the evolutionary mismatch hypothesis for breast cancer can account for ER-positive breast cancer susceptibility but not ER-negative breast cancer.
The objective of this study was to use successful quality improvement initiatives in large multis... more The objective of this study was to use successful quality improvement initiatives in large multispecialty medical groups to identify the organizational factors that were the most important to improvement. The study analyzed the most successful quality improvement initiatives from those submitted by the 24 members of the Council of Accountable Physician Practices. Twelve initiatives from 8 groups were selected that met the study criteria for large improvement for large numbers of patients. An independent group used these initiatives to identify potentially important factors and then asked key local leaders to rate the importance of these factors on a scale of 1 to 4, importance rating (1-4 scale) for each of 18 identified factors. Eighteen factors were identified and 5 stood out as ranked a 4 (Very Important) for at least 80% of the initiatives: Communication, Use of Evidence-Based Medicine, Leadership, Measurement, and Reporting. Another 7 of the 18 factors were ranked a 4 for more ...
In this paper, we describe the attributes of a comprehensive approach to breast cancer screening ... more In this paper, we describe the attributes of a comprehensive approach to breast cancer screening possible in an integrated health care system. We define an integrated health care system as one in which comprehensive preventive and medical care is provided to a defined population, by a defined panel of providers, and in which this care can be tracked using automated electronic data systems. Guided by the Pathways Conceptual Framework, it is possible to identify and systematically address (through research and interventions) the multiple predisposing, enabling, and reinforcing factors at the individual and organizational level associated with each step along the screening process. This framework is helpful as both a planning and an evaluation tool, in identifying places in the screening and follow-up process that could benefit from concerted quality improvement efforts and in guiding an evaluation of those efforts. We describe examples from research and organizational programmatic eff...
In preparation for jointly publishing official government cancer statistics, the Centers for Dise... more In preparation for jointly publishing official government cancer statistics, the Centers for Disease Control and Prevention (CDC) and the National Cancer Institute (NCI) compared incidence rates from NCI's Surveillance Epidemiology and End Results (SEER) Program and CDC's National Program of Cancer Registries (NPCR). Data for 1999 covering 78% of the US population were obtained from SEER and selected NPCR registries that met high quality data criteria. Incidence rates (per 100,000 population) were age-adjusted to the 2000 US standard population, and 95% gamma confidence intervals were estimated. NPCR rates for all sites combined were higher than SEER rates (males: NPCR 553.6, SEER 538.7; females: NPCR 420.8, SEER 412.5), but rates for specific cancer sites varied by registry program. Rates for colon cancer (males: NPCR 47.0, SEER 42.7; females: NPCR 36.5, SEER 33.8) and tobacco-related cancers were higher in NPCR than SEER. In contrast, NPCR rates were lower than SEER rates ...
Journal of the National Cancer Institute. Monographs, 1995
Current knowledge is scarce on Latino dietary practices. This study compared the dietary practice... more Current knowledge is scarce on Latino dietary practices. This study compared the dietary practices, alcohol consumption, and smoking behavior of Latinos and non-Latino whites in two randomly selected samples. Telephone surveys of adults 35-74 years of age from the Kaiser Permanente Medical Care Program (Latinos = 844; non-Latino whites = 510) and from census tract-based areas (Latinos = 806; non-Latino whites = 436) were conducted in the San Francisco Bay Area. Latino ethnicity was a significant predictor of dietary and alcohol consumption practices in multivariate logistic regression models after adjustment for sex, education, age, employment, health insurance, martial status, county of residence, and self-perceived health status. Compared with non-Latino whites, Latinos were significantly less likely to report eating vegetables and more likely to eat rice, beans, and fried foods and to drink whole milk. Less acculturated Latinos were more likely to eat fruits, rice, beans, meat, a...
Journal of the National Cancer Institute. Monographs, 1995
Difficulties in identification of Latinos from secondary datasets have often been addressed by th... more Difficulties in identification of Latinos from secondary datasets have often been addressed by the use of a Spanish surname list. To evaluate the sensitivity and specificity of the 1980 U.S. Census Spanish surname list in correctly identifying Latino ethnicity, we conducted a cross-sectional survey of adult members of a prepaid health plan in the San Francisco Bay Area. We compared a randomly selected list of 1345 Spanish surnames and 717 non-Spanish surnames from health plan membership lists with their self-identified race and ethnicity obtained by telephone interviews. The sample was stratified according to surname group (Spanish or non-Spanish), sex, age, and county. Compared with self-identified ethnicity, a Spanish surname was 88.4% sensitive in identifying Latino men and 70.4% sensitive in identifying Latina women. The non-Spanish surname list was 94.1% sensitive and 95.0% specific in correctly identifying non-Latinos. Although the overall negative predictive value was 97.8%, ...
The relation of work-related psychologic stress to spontaneous abortion was examined in a prospec... more The relation of work-related psychologic stress to spontaneous abortion was examined in a prospective study of 3,953 pregnant, employed members of a prepaid health plan in California, recruited between 1990 and 1991. Information on occupation, psychologic stress-related factors at work, social support at work, physical exertion at work, life events, pregnancy worries, and potential confounders was obtained by a telephone interview. Psychologic job stress and social support at work were assessed using an abbreviated version of the instrument developed by Karasek and colleagues (Los Angeles: University of Southern California, 1986). Results from the multivariate model indicated that, overall, stressful work was not associated with an increased risk of spontaneous abortion. However, interactions were observed between stressful work and maternal age over 32 years (p = 0.04), cigarette smoking (p = 0.02), and primigravidity (p = 0.06). Relative to the odds ratio for stressful work in you...
A case-control study of the relationship of serum cholesterol to large bowel cancer incidence was... more A case-control study of the relationship of serum cholesterol to large bowel cancer incidence was performed for 245 members of the Kaiser Permanente Medical Care Program in northern California. Five controls were obtained for each case, matched for age, sex, race, and time of examination. The mean serum cholesterol levels of cases were not significantly different from those of controls for all colorectal cancers, right-sided colon cancers, and left-sided colon cancers. Analysis by quartiles of serum cholesterol level suggested neither a direct nor an inverse relationship between serum cholesterol and large bowel cancer, and there was no evidence of a threshold value below which the risk of cancer increased. The study did not support the findings from other studies of an inverse relationship between serum cholesterol and large bowel cancer.
Uploads
Papers by Robert Hiatt