Biomedical subjects

Michael P Fay

Publications and source records attributed to Michael P Fay.

11 recordsLinked to original sources

PubMed2006

Estimating average annual percent change for disease rates without assuming constant change.

The annual percent change (APC) is often used to measure trends in disease and mortality rates, and a common estimator of this parameter uses a linear model on the log of the age-standardized rates. Under the assumption of linearity on the log scale, which is equivalent to a constant change assumption, APC can be equivalently defined in three ways as transformations of either (1) the slope of the line that runs through the log of each rate, (2) the ratio of the last rate to the first rate in the series, or (3) the geometric mean of the proportional changes in the rates over the series. When the constant change assumption fails then the first definition cannot be applied as is, while the second and third definitions unambiguously define the same parameter regardless of whether the assumption holds. We call this parameter the percent change annualized (PCA) and propose two new estimators of it. The first, the two-point estimator, uses only the first and last rates, assuming nothing about the rates in between. This estimator requires fewer assumptions and is asymptotically unbiased as the size of the population gets large, but has more variability since it uses no information from the middle rates. The second estimator is an adaptive one and equals the linear model estimator with a high probability when the rates are not significantly different from linear on the log scale, but includes fewer points if there are significant departures from that linearity. For the two-point estimator we can use confidence intervals previously developed for ratios of directly standardized rates. For the adaptive estimator, we show through simulation that the bootstrap confidence intervals give appropriate coverage.

Biometry↗

PubMed2006

Health workers' agreement in clinical description of filarial lymphedema.

Severity of lymphedema and presence of entry lesions are risk factors for acute bacterial dermatolym-phangioadenitis (ADLA) in those with filarial lymphedema. Recurrent ADLA causes acute morbidity and progression of lymphedema severity; however, there is little work assessing the ability of health workers to reliably stage disease severity and identify risk entry lesions. This knowledge is important in initiation of management and assessing interventions. We evaluated inter-rater reliability with two independent health workers rating both legs of 17 patients using a questionnaire and the Dreyer classification of lymphedema. The health workers could reliably stage lymphedema with high agreement (RMAC weighted kappa of 0.89) and identify nail, interdigital, and other skin lesions. However, there was less consistency in identifying the clinical nature of skin lesions. This indicates that the Dreyer classification can be a replicable way to stage lymphedema and a questionnaire can deliver high observer agreement on the presence of risk lesions.

Animals↗

PubMed2005

Modeling the dissemination of mammography in the United States.

OBJECTIVE: This paper presents a methodology for piecing together disparate data sources to obtain a comprehensive model for the use of mammography screening in the US population for the years 1975-2000. METHODS: Two aspects of mammography usage, the age that a woman receives her first mammography and the interval between subsequent mammograms, are modeled separately. The initial dissemination of mammography is based on cross-sectional self report data from national surveys and the interval length between screening exams is fit using longitudinal mammography registry data. RESULTS: The two aspects of mammography usage are combined to simulate screening histories for individual women that are representative of the US population. Simulated mammography patterns for the years 1994-2000 were found to be similar to observed screening patterns from the state level mammography registry for Vermont. CONCLUSIONS: The model presented gives insight into screening practices over time and provides an alternative public health measure for screening usage in the US population. The comprehensive description of mammography use from its introduction represents an important first step to understanding the impact of mammography on breast cancer incidence and mortality.

Age Distribution↗

PubMed2005

Random marginal agreement coefficients: rethinking the adjustment for chance when measuring agreement.

Agreement coefficients quantify how well a set of instruments agree in measuring some response on a population of interest. Many standard agreement coefficients (e.g. kappa for nominal, weighted kappa for ordinal, and the concordance correlation coefficient (CCC) for continuous responses) may indicate increasing agreement as the marginal distributions of the two instruments become more different even as the true cost of disagreement stays the same or increases. This problem has been described for the kappa coefficients; here we describe it for the CCC. We propose a solution for all types of responses in the form of random marginal agreement coefficients (RMACs), which use a different adjustment for chance than the standard agreement coefficients. Standard agreement coefficients model chance agreement using expected agreement between two independent random variables each distributed according to the marginal distribution of one of the instruments. RMACs adjust for chance by modeling two independent readings both from the mixture distribution that averages the two marginal distributions. In other words, both independent readings represent first a random choice of instrument, then a random draw from the marginal distribution of the chosen instrument. The advantage of the resulting RMAC is that differences between the two marginal distributions will not induce greater apparent agreement. As with the standard agreement coefficients, the RMACs do not require any assumptions about the bivariate distribution of the random variables associated with the two instruments. We describe the RMAC for nominal, ordinal and continuous data, and show through the delta method how to approximate the variances of some important special cases.

Data Interpretation, Statistical↗

PubMed2004

Estimating age conditional probability of developing disease from surveillance data.

Fay, Pfeiffer, Cronin, Le, and Feuer (Statistics in Medicine 2003; 22; 1837-1848) developed a formula to calculate the age-conditional probability of developing a disease for the first time (ACPDvD) for a hypothetical cohort. The novelty of the formula of Fay et al (2003) is that one need not know the rates of first incidence of disease per person-years alive and disease-free, but may input the rates of first incidence per person-years alive only. Similarly the formula uses rates of death from disease and death from other causes per person-years alive. The rates per person-years alive are much easier to estimate than per person-years alive and disease-free. Fay et al (2003) used simple piecewise constant models for all three rate functions which have constant rates within each age group. In this paper, we detail a method for estimating rate functions which does not have jumps at the beginning of age groupings, and need not be constant within age groupings. We call this method the mid-age group joinpoint (MAJ) model for the rates. The drawback of the MAJ model is that numerical integration must be used to estimate the resulting ACPDvD. To increase computational speed, we offer a piecewise approximation to the MAJ model, which we call the piecewise mid-age group joinpoint (PMAJ) model. The PMAJ model for the rates input into the formula for ACPDvD described in Fay et al (2003) is the current method used in the freely available DevCan software made available by the National Cancer Institute.

Journal Article↗

PubMed2004

Trends in drinking among Americans age 18 and younger: 1975-2002.

BACKGROUND: Although changes over time in the prevalence of drinking by youth under 18 have been previously reported, we present results based on data from multiple surveys, using recently developed software for trend analyses. In this study, we applied joinpoint statistical methodology to three national data sets to analyze trends in drinking by youth, age 18 and under, for the period 1975 to 2002. METHODS: Information was obtained from three national data sets, Monitoring the Future for the years 1975 to 2002, the Youth Risk Behavior Survey for the years 1991 to 2001, and the National Household Survey on Drug Abuse for 1979, 1985, and 1991 to 2001. Approximately 80,000 persons between 12 and 18 were included in the most recent survey years. The alcohol consumption measures examined over time were any use of alcohol, consumption of five or more drinks on one occasion, and daily consumption. RESULTS: Alcohol consumption by 8th, 10th and 12th graders decreased substantially since the 1970s according to joinpoint trend analyses. It remains disturbingly high, however, according to data from three national surveys (e.g., 12.4% of 8 and 28.6% of 12th graders drinking five or more drinks in a row in the past 2 weeks), although prevalence rates have been relatively stable for the last 5 to 10 years. CONCLUSIONS: Since the early 1990s, rates of drinking by youth under 18 remained relatively stable according to the Youth Risk Behavior Survey and National Household Survey on Drug Abuse and moved up and then down according to Monitoring the Future, underscoring the need for continued surveillance and enhanced understanding of this long-standing problem.

Adolescent↗

PubMed2004

Comparability of segmented line regression models.

Segmented line regression models, which are composed of continuous linear phases, have been applied to describe changes in rate trend patterns. In this article, we propose a procedure to compare two segmented line regression functions, specifically to test (i) whether the two segmented line regression functions are identical or (ii) whether the two mean functions are parallel allowing different intercepts. A general form of the test statistic is described and then the permutation procedure is proposed to estimate the p-value of the test. The permutation test is compared to an approximate F-test in terms of the p-value estimation and the performance of the permutation test is studied via simulations. The tests are applied to compare female lung cancer mortality rates between two registry areas and also to compare female breast cancer mortality rates between two states.

Biometry↗

PubMed2004

Weight change and the risk of late-onset breast cancer in the original Framingham cohort.

OBJECTIVE: Adult weight gain has been associated with a twofold risk of postmenopausal breast cancer. Data are limited regarding whether weight gain at specific periods of marked changes in estrogen- and insulin-related hormones have different risk associations. This study assesses the relation of adult weight change overall and at specific, hormonally relevant times with diagnosis of a first breast cancer after age 55 (late onset). METHODS: Framingham study data were used to assess premenopausal (25-44 yr), perimenopausal (45-55 yr), postmenopausal (after 55 yr), and adult lifetime (from 25 yr) weight change in relation to late-onset breast cancer in 2,873 women followed for up to 48 yr, with 206 late-onset breast cancers. RESULTS: Adult lifetime weight gain was associated with an increased risk of late-onset breast cancer (P trend = 0.046). Weight gain during specific time periods was not associated with breast cancer. Data suggested a possible decreased risk of breast cancer with weight loss from ages 25 to 44 and 45 to 55 yr (relative risk = 0.4 [0.2-1.2] and 0.5 [0.3-0.9], respectively). CONCLUSION: These data confirm prior reports of an association between adult lifetime weight gain and increased risk of late-onset breast cancer and support current recommendations to avoid adult weight gain.

Adult↗

PubMed2003

Age-conditional probabilities of developing cancer.

We propose an estimator of the probability of developing a disease in a given age range, conditional on never having developed the disease prior to the beginning of the age range. Our estimator improves the one described by Wun, Merrill and Feuer ( Lifetime Data Analysis 1998; 4, 169-186) that is currently used by the U.S. National Cancer Institute for the SEER Cancer Statistics Review. Both estimators use cross-sectional disease rates and provide an interpretation of these rates in terms of the age-conditional probability of developing disease in a hypothetical cohort. The difficulty of this problem is that rates are not available per person-years alive and disease free, but only per person-years alive. Wun et al. used ad hoc methods to handle this problem which did not properly account for competing risks, did not provide a measure of variability, and only allowed age ranges using prespecified 5-year age intervals. Here we solve the problem under a unified competing risks framework, which allows the calculation of the age-conditional probabilities for any age range. We generalize gamma confidence intervals to apply to our new statistic. Although our new method provides estimates which are numerically similar to that of Wun et al., this paper provides a comprehensive theoretical basis for estimation and inference about the age-conditional probability of developing a disease.

Age of Onset↗

PubMed2002

Impact of reporting delay and reporting error on cancer incidence rates and trends.

BACKGROUND: Cancer incidence rates and trends are a measure of the cancer burden in the general population. We studied the impact of reporting delay and reporting error on incidence rates and trends for cancers of the female breast, colorectal, lung/bronchus, prostate, and melanoma. METHODS: Based on statistical models, we obtained reporting-adjusted (i.e., adjusted for both reporting delay and reporting error) case counts for each diagnosis year beginning in 1981 using reporting information for patients diagnosed with cancer in 1981-1998 from nine cancer registries that participate in the Surveillance, Epidemiology, and End Results (SEER) program. Joinpoint linear regression was used for trend analysis. All statistical tests are two-sided. RESULTS: Initial incidence case counts (i.e., after the standard 2-year delay) accounted for only 88%-97% of the estimated final counts; it would take 4-17 years for 99% or more of the cancer cases to be reported. The percent change between reporting-adjusted and unadjusted cancer incidence rates for the 1998 diagnosis year ranged from 3% for colorectal cancers to 14% for melanoma in whites and for prostate cancer in black males. Reporting-adjusted current incidence trends for breast cancer and lung/bronchus cancer in white females showed statistically significant increases (estimated annual percent change [EAPC] = 0.6%, 95% confidence interval [CI] = 0.1% to 1.2%) and 1.2%, 95% CI = 0.7% to 1.6%, respectively), whereas trends for these cancers using unadjusted incidence rates were not statistically significantly different from zero (EAPC = 0.4%, 95% CI = -0.1% to 0.9% and 0.5%, 95% CI = -0.1% to 1.1%, respectively). Reporting-adjusted melanoma incidence rates for white males showed a statistically significant increase since 1981 (EAPC = 4.1%, 95% CI = 3.8% to 4.4%) in contrast to the unadjusted incidence rate, which was most consistent with a flat or downward trend (EAPC = -4.2%, 95% CI = -11.1% to 3.3%) after 1996. CONCLUSIONS: Reporting-adjusted cancer incidence rates are valuable in precisely determining current cancer incidence rates and trends and in monitoring the timeliness of data collection. Ignoring reporting delay and reporting error may produce downwardly biased cancer incidence trends, particularly in the most recent diagnosis years.

Adult↗

PubMed2002

Costs of treatment for elderly women with early-stage breast cancer in fee-for-service settings.

PURPOSE: This study provides population-based estimates of the treatment costs for elderly women with early-stage breast cancer, with emphasis on costs of modified radical mastectomy (MRM) compared with breast-conserving surgery (BCS) and radiation therapy (RT). PATIENTS AND METHODS: Women with breast cancer from the Surveillance, Epidemiology, and End Results cancer registries were linked with their Medicare claims, 1990 through 1998. Each claim was assigned to an initial, continuing, or terminal care phase after a cancer diagnosis. Mean monthly phase-specific costs were determined for all health care and for treatment related only to cancer. Cumulative long-term costs of care that accrue during a women's remaining lifetime were calculated by treatment group. RESULTS: Initial care costs for the 6 months after diagnosis for women who underwent BCS with RT were approximately $450 per month higher than for women with MRM. During the continuing-care phase, costs for women undergoing BCS with RT were significantly less expensive than for MRM cases. The two groups had similar costs in the terminal-care phase. Assuming the same survival distributions, long-term costs for women undergoing BCS with RT were not statistically different than for women undergoing MRM. CONCLUSION: Although mastectomy was less costly in the initial phase, the lifetime costs of BCS with RT and mastectomy were equivalent. Thus, women's preferences, resources to cover out-of-pocket costs, and life situations should be the major factors addressed in shared decision making about treatment options.

Aged↗