Assessment of Birth Defects and Cancer Risk in Children Conceived via In Vitro Fertilization in the US.

Importance: Children with birth defects have a greater risk of developing cancer, but this association has not yet been evaluated in children conceived with in vitro fertilization (IVF). Objective: To assess whether the association between birth defects and cancer is greater in children conceived via IVF compared with children conceived naturally. Design, Setting, and Participants: This cohort study of live births, birth defects, and cancer from Massachusetts, New York, North Carolina, and Texas included 1 000 639 children born to fertile women and 52 776 children conceived via IVF (using autologous oocytes and fresh embryos) during 2004-2016 in Massachusetts and North Carolina, 2004-2015 in New York, and 2004-2013 in Texas. Children were followed up for an average of 5.7 years (6 008 985 total person-years of exposure). Data analysis was conducted from April 1 to August 31, 2020. Exposures: Conception by IVF for state residents who gave birth to liveborn singletons during the study period. Birth defect diagnoses recorded by statewide registries. Main Outcomes and Measures: Cancer diagnosis as recorded by state cancer registries. Cox proportional hazards regression models were used to estimate hazard ratios (HRs) and 95% CIs for birth defect-cancer associations separately in fertile and IVF groups. Results: A total of 1 000 639 children (51.3% boys; 69.7% White; and 38.3% born between 2009-2012) were in the fertile group and 52 776 were in the IVF group (51.3% boys; 81.3% White; and 39.6% born between 2009-2012). Compared with children without birth defects, cancer risks were higher among children with a major birth defect in the fertile group (hazard ratio [HR], 3.15; 95% CI, 2.40-4.14) and IVF group (HR, 6.90; 95% CI, 3.73-12.74). The HR of cancer among children with a major nonchromosomal defect was 2.07 (95% CI, 1.47-2.91) among children in the fertile group and 4.04 (95% CI, 1.86-8.77) among children in the IVF group. The HR of cancer among children with a chromosomal defect was 15.45 (95% CI, 10.00-23.86) in the fertile group and 38.91 (95% CI, 15.56-97.33) in the IVF group. Conclusions and Relevance: This study found that among children with birth defects, those conceived via IVF were at greater risk of developing cancer compared with children conceived naturally.

Third grade academic achievement among children conceived with the use of in vitro fertilization: a population-based study in Texas.

OBJECTIVE: To evaluate if there are differences in standardized testing results at the end of third grade between children conceived with the use of in vitro fertilization (IVF) and those conceived spontaneously. DESIGN: Retrospective population-based cohort. SETTING: Texas public school system. PATIENT(S): Singleton and twin children 8-9 years of age who took the third-grade public school standardized testing in Texas from 2012 to 2018. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Standardized testing in reading and mathematics. RESULT(S): After exclusions, there were 6,970 IVF and 12,690 non-IVF children with reading scores and 6,973 IVF and 12,729 non-IVF children with mathematics scores. IVF children scored significantly higher in reading (singletons: 1,543 ± 2 vs. 1,525 ± 1; twins: 1,534 ± 2 vs. 1,504 ± 5 [mean ± SE]), and mathematics (singletons: 1,566 ± 2 vs. 1,550 ± 1; twins: 1,557 ± 2 vs. 1,529 ± 5). Children of mothers ≥30 years of age scored consistently higher than children of mothers 18-29 years of age. The differences were of similar magnitude between IVF and control children for older ages, but not significant for IVF. Within the IVF group, there were no significant differences between children born from fresh versus froze-thawed embryos. CONCLUSION(S): Children of ages 8-9 years who were conceived with the use of IVF performed as well on third-grade reading and math assessments as their counterparts who were conceived spontaneously. We also found consistent racial and ethnic differences, gender differences, and beneficial effects of older maternal age. Because we were not able to adjust adequately for socioeconomic status and other confounding factors, which may explain some of the observed differences, we conclude that there is no negative effect of IVF conception on academic achievement in third grade.

Assisted reproductive technology use and outcomes among women with a history of cancer.

STUDY QUESTION: How do the assisted reproductive technology (ART) outcomes of women presenting for ART after cancer diagnosis compare to women without cancer? SUMMARY ANSWER: The likelihood of a live birth after ART among women with prior cancer using autologous oocytes is reduced and varies by cancer diagnosis but is similar to women without cancer when donor oocytes are used. WHAT IS KNOWN ALREADY: Premenopausal patients faced with a cancer diagnosis frequently present for fertility preservation. STUDY DESIGN, SIZE, DURATION: Population-based cohort study of women treated with ART in NY, TX and IL, USA. PARTICIPANTS/MATERIALS, SETTING, METHODS: Women with their first ART treatment between 2004 and 2009 were identified from the Society for Assisted Reproductive Technology Clinic Outcome Reporting System database and linked to their respective State Cancer Registries based on name, date of birth and social security number. Years were rounded, i.e. year 1 = 6-18 months before treatment. This study used reports of cancer from 5 years, 6 months prior to treatment until 6 months after first ART treatment. Women who only presented for embryo banking were omitted from the analysis. The likelihood of pregnancy and of live birth with ART using autologous oocytes was modeled using logistic regression, with women without prior cancer as the reference group, adjusted for woman's age, parity, cumulative FSH dosage, infertility diagnosis, number of diagnoses, number of ART cycles, State of residency and year of ART treatment. Results of the modeling are reported as adjusted odds ratios (AORs) and (95% confidence intervals). MAIN RESULTS AND THE ROLE OF CHANCE: The study population included 53 426 women; 441 women were diagnosed with cancer within 5 years prior to ART cycle start. Mean (±SD) age at cancer diagnosis was 33.4 ± 5.7 years; age at start of ART treatment was 34.9 ± 5.8 for women with cancer compared with 35.3 ± 5.3 years for women without cancer (P = 0.03). Live birth rates among women using autologous oocytes differed substantially by cancer status (47.7% without cancer versus 24.7% with cancer, P < 0.0001), and cancer diagnosis (ranging from 53.5% for melanoma to 14.3% for breast cancer, P < 0.0001. The live birth rates among women using donor oocytes did not vary significantly by cancer status (60.4% for women with any cancer versus 64.5% for women without cancer), or by cancer diagnosis (ranging from 57.9% for breast cancer to 63.6% for endocrine cancer). Women with breast cancer make up about one-third of all cancers in this cohort. Among women with breast cancer, 2.8% of the 106 women who underwent ART within 6 months of being diagnosed with cancer used donor oocytes compared with 34.8% of the 46 women who received ART treatment a longer time after being diagnosed with cancer (P < 0.0001). We conjecture that the former group were either unaware that they had cancer or decided to undergo ART therapy prior to cancer treatment. However, their live birth rate was only 11.7% compared with 28.8%, the overall live birth rate for all women with cancer using autologous oocytes (P < 0.0001). The live birth rate for women diagnosed with breast cancer more than 6 months before ART (23.3%) did not differ significantly from the overall live birth rate for cancer (P = 0.49). If this difference is substantiated by a larger study, it would indicate a negative effect of severe recent illness itself on ART success, rather than the poor outcome being only related to the destructive effects of chemotherapies on ovarian follicles. Alternatively, because of the short time difference between cancer diagnosis and ART treatment, these pre-existing cancers may have been detected due to the increased medical surveillance during ART therapy. In women who only used autologous oocytes, women with prior cancers were significantly less likely to become pregnant and to have a live birth than those without cancer (adjusted odds ratio (AOR): 0.34, [95% confidence interval (CI): 0.27, 0.42] and 0.36 [0.28, 0.46], respectively). This was also evident with specific cancer diagnoses: breast cancer (0.20 [0.13, 0.32] and 0.19 [0.11, 0.30], respectively), cervical cancer (0.36 [0.15, 0.87] and 0.33 [0.13, 0.84], respectively) and all female genital cancers (0.49 [0.27, 0.87] and 0.47 [0.25, 0.86], respectively). Of note, among women with cancer who became pregnant, their likelihood of having a live birth did not differ significantly from women without cancer (85.8 versus 86.7% for women using autologous oocytes, and 85.3 versus 86.9% for women using donor oocytes). LIMITATIONS, REASONS FOR CAUTION: Women may not have been residents of the individual States for the entire 5-year pre-ART period, and therefore some cancers may not have been identified through this linkage. As a result, the actual observed number of cancers may be an underestimate. In addition, the overall prevalence is low due to the age distributions. Also, because we restricted the pre-ART period to 5 years prior, we would not have identified women who were survivors of early childhood cancers (younger than age 13 years at cancer diagnosis), or who had ART more than 5 years after being diagnosed with cancer. Additional analyses are currently underway evaluating live birth outcomes after embryo banking among women with cancer prior to ART, cycles which were excluded from the analyses in this paper. Future studies are planned which will include more States, as well as linkages to vital records to obtain information on spontaneous conceptions and births, to further clarify some of the issues raised in this analysis. WIDER IMPLICATIONS OF THE FINDINGS: Since the live birth rates using donor oocytes were not reduced in women with a prior cancer, but were reduced with autologous cycles, this suggests that factors acting in the pre- or peri-conceptional periods may be responsible for the decline. STUDY FUNDING/COMPETING INTERESTS: The study was funded by grant R01 CA151973 from the National Cancer Institute, National Institutes of Health, USA. B.L. is a research consultant for the Society for Assisted Reproductive Technology. All other authors report no conflict of interest.

The Michigan Model for Coronary Heart Disease in Type 2 Diabetes: Development and Validation.

OBJECTIVES: The aim of this study was to develop and validate a computer simulation model for coronary heart disease (CHD) in type 2 diabetes mellitus (T2DM) that reflects current medical and surgical treatments. RESEARCH DESIGN AND METHODS: We modified the structure of the CHD submodel in the Michigan Model for Diabetes to allow for revascularization procedures before and after first myocardial infarction, for repeat myocardial infarctions and repeat revascularization procedures, and for congestive heart failure. Transition probabilities that reflect the direct effects of medical and surgical therapies on outcomes were derived from the literature and calibrated to recently published population-based epidemiologic studies and randomized controlled clinical trials. Monte Carlo techniques were used to implement a discrete-state and discrete-time multistate microsimulation model. Performance of the model was assessed using internal and external validation. Simple regression analysis (simulated outcome=b(0)+b(1)×published outcome) was used to evaluate the validation results. RESULTS: For the 21 outcomes in the six studies used for internal validation, R(2) was 0.99, and the slope of the regression line was 0.98. For the 16 outcomes in the five studies used for external validation, R(2) was 0.81, and the slope was 0.84. CONCLUSIONS: Our new computer simulation model predicted the progression of CHD in patients with T2DM and will be incorporated into the Michigan Model for Diabetes to assess the cost-effectiveness of alternative strategies to prevent and treat T2DM.

Live birth rates and birth outcomes by diagnosis using linked cycles from the SART CORS database.

PURPOSE: This study uses linked cycles of assisted reproductive technology (ART) to examine cumulative live birth rates, birthweight, and length of gestation by diagnostic category. METHODS: We studied 145,660 women with 235,985 ART cycles reported to the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System during 2004-2010. ART cycles were linked to individual women by name, date of birth, social security number, partner's name, and sequence of ART treatments. The study population included the first four autologous oocyte cycles for women with a single diagnosis of male factor, endometriosis, ovulation disorders, diminished ovarian reserve, or unexplained infertility. Live birth rates were calculated per cycle, per cycle number (1-4), and cumulatively. Birthweight and length of gestation were calculated for singleton births. RESULTS: Within each diagnosis, live birth rates were highest in the first cycle and declined with successive cycles. Women with diminished ovarian reserve had the lowest live birth rate (cumulative rate of 28.3 %); the live birth rate for the other diagnoses were very similar (cumulative rates from 62.1 % to 65.7 %). Singleton birthweights and lengths of gestation did not differ substantially across diagnoses, ranging from 3,112 to 3,286 g and 265 to 270 days, respectively. These outcomes were comparable with national averages for singleton births in the United States (3,296 g and 271 days). CONCLUSION: Women with the diagnosis of diminished ovarian reserve had substantially lower live birth rates. However, singleton birthweights and lengths of gestation outcomes were similar across all other diagnoses.

Medical costs associated with type 2 diabetes complications and comorbidities.

OBJECTIVES: To estimate the direct medical costs associated with type 2 diabetes, its complications, and its comorbidities among U.S. managed care patients. STUDY DESIGN: Data were from patient surveys, chart reviews, and health insurance claims for 7109 people with type 2 diabetes from 8 health plans participating in the Translating Research Into Action for Diabetes (TRIAD) study between 1999 and 2002. METHODS: A generalized linear regression model was developed to estimate the association of patients' demographic characteristics, tobacco use status, treatments, related complications, and comorbidities with medical costs. RESULTS: The mean annualized direct medical cost was $2465 for a white man with type 2 diabetes who had been diagnosed fewer than 15 years earlier, was treated with oral medication or diet alone, and had no complications or comorbidities. We found annualized medical costs to be 10% to 50% higher for women and for patients whose diabetes had been diagnosed 15 or more years earlier, who used tobacco, who were being treated with insulin, or who had several other complications. Coronary heart disease, congestive heart failure, hemiplegia, and amputation were each associated with 70% to 150% higher costs. Costs were approximately 300% higher for end-stage renal disease treated with dialysis and approximately 500% higher for end-stage renal disease with kidney transplantation. CONCLUSIONS: Most medical costs incurred by patients with type 2 diabetes are related to complications and comorbidities. Our cost estimates can help when determining the most cost-effective interventions to prevent complications and comorbidities.

Effectiveness and cost-effectiveness of diabetes prevention among adherent participants.

OBJECTIVES: We report the 10-year effectiveness and within-trial cost-effectiveness of the Diabetes Prevention Program (DPP) and its Outcomes Study (DPPOS) interventions among participants who were adherent to the interventions. STUDY DESIGN: DPP was a 3-year randomized clinical trial followed by 7 years of open-label modified intervention follow-up. METHODS: Data on resource utilization, cost, and quality of life were collected prospectively. Economic analyses were performed from health system and societal perspectives. Lifestyle adherence was defined as achieving and maintaining a 5% reduction in initial body weight, and metformin adherence as taking metformin at 80% of study visits. RESULTS: The relative risk reduction was 49.4% among adherent lifestyle participants and 20.8% among adherent metformin participants compared with placebo. Over 10 years, the cumulative, undiscounted, per capita direct medical costs of the interventions, as implemented during the DPP, were greater for adherent lifestyle participants ($4810) than adherent metformin participants ($2934) or placebo ($768). Over 10 years, the cumulative, per capita non-interventionrelated direct medical costs were $4250 greater for placebo participants compared with adherent lifestyle participants and $3251 greater compared with adherent metformin participants. The cumulative quality-adjusted life-years (QALYs) accrued over 10 years were greater for lifestyle (6.80) than metformin (6.74) or placebo (6.67). Without discounting, from a modified societal perspective (excluding participant time) and a full societal perspective (including participant time), lifestyle cost < $5000 per QALY-gained and metformin was cost saving compared with placebo. CONCLUSIONS: Over 10 years, lifestyle intervention and metformin were cost-effective or cost saving compared with placebo. These analyses confirm that lifestyle and metformin represent a good value for money.

Racial and ethnic disparities in assisted reproductive technology pregnancy and live birth rates within body mass index categories.

OBJECTIVE: To evaluate the effect of maternal race and ethnicity within body mass index (BMI) categories on assisted reproduction technology (ART) pregnancy and live birth rates. DESIGN: Historical cohort study. SETTING: Clinic-based data. PATIENT(S): 31,672 ART embryo transfers from the Society for Assisted Reproductive Technology Clinic Outcomes Reporting System for 2007, limited to women with documented race, ethnicity, height, and weight, with women grouped as white, Asian, Hispanic, or black and by BMI. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Failure to achieve a clinical intrauterine gestation and failure to achieve a live birth as adjusted odds ratios within BMI categories overall with normal-weight women as the reference group, and by race and ethnicity with white women as the reference group. RESULT(S): Failure to achieve a clinical intrauterine gestation was significantly more likely among obese women overall (1.22), normal-weight and obese Asian women (1.36 and 1.73, respectively), normal-weight Hispanic women (1.21), and overweight and obese black women (1.34 and 1.47, respectively). Among women who did conceive, failure to achieve a live birth was significantly more likely among overweight and obese women overall (1.16 and 1.27, respectively), overweight and obese Asian women (1.56 and 2.20, respectively) and Hispanic women (1.57 and 1.76, respectively), and normal-weight and obese black women (1.45 and 1.84, respectively). CONCLUSION(S): These findings indicate significant disparities in pregnancy and live birth rates by race and ethnicity, even within BMI categories.

Thiazolidinediones, cardiovascular disease and cardiovascular mortality: translating research into action for diabetes (TRIAD).

BACKGROUND: Studies have associated thiazolidinedione (TZD) treatment with cardiovascular disease (CVD) and questioned whether the two available TZDs, rosiglitazone and pioglitazone, have different CVD risks. We compared CVD incidence, cardiovascular (CV), and all-cause mortality in type 2 diabetic patients treated with rosiglitazone or pioglitazone as their only TZD. METHODS: We analyzed survey, medical record, administrative, and National Death Index (NDI) data from 1999 through 2003 from Translating Research Into Action for Diabetes (TRIAD), a prospective observational study of diabetes care in managed care. Medications, CV procedures, and CVD were determined from health plan (HP) administrative data, and mortality was from NDI. Adjusted hazard rates (AHR) were derived from Cox proportional hazard models adjusted for age, sex, race/ethnicity, income, history of diabetic nephropathy, history of CVD, insulin use, and HP. RESULTS: Across TRIAD's 10 HPs, 1,815 patients (24%) filled prescriptions for a TZD, 773 (10%) for only rosiglitazone, 711 (10%) for only pioglitazone, and 331 (4%) for multiple TZDs. In the seven HPs using both TZDs, 1,159 patients (33%) filled a prescription for a TZD, 564 (16%) for only rosiglitazone, 334 (10%) for only pioglitazone, and 261 (7%) for multiple TZDs. For all CV events, CV, and all-cause mortality, we found no significant difference between rosiglitazone and pioglitazone. CONCLUSIONS: In this relatively small, prospective, observational study, we found no statistically significant differences in CV outcomes for rosiglitazone- compared to pioglitazone-treated patients. There does not appear to be a pattern of clinically meaningful differences in CV outcomes for rosiglitazone- versus pioglitazone-treated patients.

Practice patterns and outcomes with the use of single embryo transfer in the United States.

OBJECTIVE: To evaluate factors associated with the use of elective single embryo transfer (eSET) and its effect on assisted reproductive technology (ART) outcome. DESIGN: Historical cohort. SETTING: Clinic-based data. PATIENT(S): A total of 69,028 ART cycles of autologous fresh embryo transfers with additional embryos cryopreserved during the same cycle performed during 2004-06 and reported to the Society for Assisted Reproductive Technology Clinic Outcome Reporting System database. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Factors associated with the number of embryos transferred, and the odds of pregnancy, live birth, and multiple-infant live birth by number of embryos transferred as adjusted odds ratios (AORs). RESULT(S): Single embryo transfer was used more with uterine factor (AOR 1.76) and less with male factor, endometriosis, or tubal factor (AOR 0.81, 0.72, 0.83, respectively). Compared with women aged <30 years, eSET was used less among women aged 35-39 years and > or =40 years (AOR 0.74 and 0.39, respectively). Compared with White women, eSET was used more with Asian (AOR 1.52) and less with Black or Hispanic women (AOR 0.73 and 0.67, respectively). Compared with eSET, the likelihood of pregnancy, live birth, or multiple-infant live birth was more likely with two embryos (AOR 1.33, 1.34, and 27.4, respectively). CONCLUSION(S): Elective SET, used more for younger women with specific diagnoses, is associated with slightly reduced likelihood of a live birth but much reduced likelihood of multiples.

Racial and ethnic disparities in assisted reproductive technology outcomes in the United States.

OBJECTIVE: To evaluate ethnic differences in assisted reproductive technology (ART) outcomes in the United States. DESIGN: Historical cohort study. SETTING: Clinic-based data. PATIENT(S): A total of 139,027 ART cycles from the Society for Assisted Reproductive Technology Clinic Outcome Reporting System online database for 2004-2006, limited to white, Asian, black, and Hispanic women. INTERVENTION(S): None. MAIN OUTCOME MEASURE(S): Logistic regression was used to model the odds of pregnancy and live birth; among singletons and twins, the odds of preterm birth and fetal growth restriction. Results are presented as adjusted odds ratios, with white women as the reference group. RESULT(S): The odds of pregnancy were reduced for Asians (0.86), and the odds of live birth were reduced for all groups: Asian (0.90), black (0.62), and Hispanic (0.87) women. Among singletons, moderate and severe growth restriction were increased for all infants in all three minority groups (Asians [1.78, 2.05]; blacks [1.81, 2.17]; Hispanics [1.36, 1.64]), and preterm birth was increased among black (1.79) and Hispanic women (1.22). Among twins, the odds for moderate growth restriction were increased for infants of Asian (1.30) and black women (1.97), and severe growth restriction was increased among black women (3.21). The odds of preterm birth were increased for blacks (1.64) and decreased for Asians (0.70). CONCLUSION(S): There are significant disparities in ART outcomes according to ethnicity.

The impact of fecal and urinary incontinence on quality of life 6 months after childbirth.

OBJECTIVE: The objective of the study was to investigate the impact of postpartum fecal incontinence (FI) and urinary incontinence (UI) on quality of life (QOL). STUDY DESIGN: Seven hundred fifty-nine primiparous women in the Childbirth and Pelvic Symptoms study were interviewed 6 months postpartum. FI and UI were assessed with validated questionnaires. We measured QOL with SF-12 summary scores, health utility index score (a measure of self-rated overall health), and the modified Manchester Health Questionnaire. RESULTS: Women with FI had worse self-rated health utility index scores (85.1 +/- 9.8 vs 88.0 +/- 11.6, P = .02) and Medical Outcomes Study Short Form Health Survey (SF-12) mental summary scores (46.8 +/- 9.2 vs 51.1 +/- 8.7, P < .0001) than women without FI or flatal incontinence. Women with UI had worse SF-12 mental summary scores (48.3 +/- 9.8 vs 51.6 +/- 7.8, P < .01) and self-rated health utility index scores (84.1 +/- 12.5 vs 88.7 +/- 10.1, P < .01) than women without UI. Women with both FI and UI had the lowest SF-12 mental summary scores (44.5 +/- 9.0). CONCLUSION: Six months after delivery, women experiencing FI or UI reported negative effects on health-related QOL. FI and UI together have a greater impact than either condition alone.

Recruitment of women research participants: the Women's Health Registry at the University of Michigan.

OBJECTIVE: The goal was to develop the Women's Health Registry, a research participant database that prospectively collects detailed information on potential research subjects to assist in linking them with open research protocols and to assess investigator use and satisfaction with this Registry. METHODS: The Women's Health Registry was launched in 1999. Women aged > or =18 years were recruited to enroll in a database of women with interest in research participation and to complete a health questionnaire. Women's health researchers with IRB-approved projects were encouraged to apply for access to the Registry participants. In 2003, the first 15 investigators to use the Women's Health Registry were asked to participate in a standardized open-ended interview to assess investigator satisfaction with this recruitment tool. RESULTS: The Women's Health Registry is currently populated with 2436 women: 36.8% aged 18-34, 39.9% aged 35-54, 16.8% aged 55-69, and 6.4% aged > or =70 years. Of these women, 84% are Caucasian and 8.5% are African American. Structured interviews with 13 of the 15 investigators contacted revealed that 36.4% of the total subject enrollment recruited by these investigators was recruited from the Women's Health Registry. In addition, Registry participants were more likely to enroll in their research protocols than women contacted through other methods. Most of the investigators' expectations from the Women's Health Registry were met, except for access to menopausal women. CONCLUSIONS: The Women's Health Registry was successfully developed, and the goal of linking women with appropriate protocols was met with significant investigator satisfaction.

A discrete-state discrete-time model using indirect observation.

This research was motivated by a desire to model the progression of a chronic disease through various disease stages when data are not available to directly estimate all the transition parameters in the model. This is a common occurrence when time and expense make it unfeasible to follow a single cohort to estimate all the transition parameters. One difficulty of developing a model of chronic disease progression from such data is that the available studies often do not include the transitions of interest. For example, in our model of diabetic nephropathy, many clinical studies did not differentiate between patients without nephropathy and those who had microalbuminuria (a pre-clinical stage of nephropathy). Another difficulty was a lack of data to directly estimate parameters of interest. We consider models which can accommodate such difficulties. In this paper we consider the problem of estimating parameters of a discrete-time Markov process when longitudinal data describing the entire process are not available. First, we present a likelihood approach to estimate parameters of a discrete-time Markov model. Next, we use simulation to investigate the finite-sample behaviour of our approach. Finally, we present two examples: a model of diabetic nephropathy and a model of cardiovascular disease in diabetes.

A computer simulation model of diabetes progression, quality of life, and cost.

OBJECTIVE: To develop and validate a comprehensive computer simulation model to assess the impact of screening, prevention, and treatment strategies on type 2 diabetes and its complications, comorbidities, quality of life, and cost. RESEARCH DESIGN AND METHODS: The incidence of type 2 diabetes and its complications and comorbidities were derived from population-based epidemiologic studies and randomized, controlled clinical trials. Health utility scores were derived for patients with type 2 diabetes using the Quality of Well Being-Self-Administered. Direct medical costs were derived for managed care patients with type 2 diabetes using paid insurance claims. Monte Carlo techniques were used to implement a semi-Markov model. Performance of the model was assessed using baseline and 4- and 10-year follow-up data from the older-onset diabetic population studied in the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR). RESULTS: Applying the model to the baseline WESDR population with type 2 diabetes, we predicted mortality to be 51% at 10 years. The prevalences of stroke and myocardial infarction were predicted to be 18 and 19% at 10 years. The prevalences of nonproliferative diabetic retinopathy, proliferative retinopathy, and macular edema were predicted to be 45, 16, and 18%, respectively; the prevalences of microalbuminuria, proteinuria, and end-stage renal disease were predicted to be 19, 39, and 3%, respectively; and the prevalences of clinical neuropathy and amputation were predicted to be 52 and 5%, respectively, at 10 years. Over 10 years, average undiscounted total direct medical costs were estimated to be USD $53,000 per person. Among survivors, the average utility score was estimated to be 0.56 at 10 years. CONCLUSIONS: Our computer simulation model accurately predicted survival and the cardiovascular, microvascular, and neuropathic complications observed in the WESDR cohort with type 2 diabetes over 10 years. The model can be used to predict the progression of diabetes and its complications, comorbidities, quality of life, and cost and to assess the relative effectiveness, cost-effectiveness, and cost-utility of alternative strategies for the prevention and treatment of type 2 diabetes.

The cost of twin pregnancy: maternal and neonatal factors.

OBJECTIVE: The purpose of this study was to evaluate factors affecting birth charges in twin pregnancies. STUDY DESIGN: Clinical and financial data were obtained on 1486 twin pregnancies delivered between 1995 to 2002 at medical centers in Maryland, Florida, Michigan, and South Carolina. Maternal and neonatal length of stay (LOS) and charges were modeled by gestational age and other risk factors using a general linear model. RESULTS: Maternal and infant birth admission LOS and charges increased significantly with a decline in gestational age. Maternal LOS and charges were also significantly increased by cesarean delivery and preeclampsia. Newborn LOS and charges increased significantly by monochorionicity and slowed growth between 20 to 28 weeks. For mother and infants, the shortest LOS and lowest birth charges were at 37 to 38 weeks. CONCLUSION: These findings reflect the substantial maternal and neonatal morbidity associated with twin pregnancies, and demonstrate that 37 to 38 weeks is their optimal gestation.

The direct medical cost of type 2 diabetes.

OBJECTIVE: To describe the direct medical costs associated with type 2 diabetes, as well as its treatments, complications, and comorbidities. RESEARCH DESIGN AND METHODS: We studied a random sample of 1,364 subjects with type 2 diabetes who were members of a Michigan health maintenance organization. Demographic characteristics, duration of diabetes, diabetes treatments, glycemic control, complications, and comorbidities were assessed by surveys and medical chart reviews. Annual resource utilization and costs were assessed using health insurance claims. The log-transformed annual direct medical costs were fitted by multiple linear regression to indicator variables for demographics, treatments, glycemic control, complications, and comorbidities. RESULTS: The median annual direct medical costs for subjects with diet-controlled type 2 diabetes, BMI 30 kg/m(2), and no microvascular, neuropathic, or cardiovascular complications were 1,700 dollars for white men and 2,100 dollars for white women. A 10-kg/m(2) increase in BMI, treatment with oral antidiabetic or antihypertensive agents, diabetic kidney disease, cerebrovascular disease, and peripheral vascular disease were each associated with 10-30% increases in cost. Insulin treatment, angina, and MI were each associated with 60-90% increases in cost. Dialysis was associated with an 11-fold increase in cost. CONCLUSIONS: Insulin treatment and diabetes complications have a substantial impact on the direct medical costs of type 2 diabetes. The estimates presented in this model may be used to analyze the cost-effectiveness of interventions for type 2 diabetes.

Shape-invariant modeling of circadian rhythms with random effects and smoothing spline ANOVA decompositions.

Medical studies often collect physiological and/or psychological measurements over time from multiple subjects, to study dynamics such as circadian rhythms. Under the assumption that the expected response functions of all subjects are the same after shift and scale transformations, shape-invariant models have been applied to analyze this kind of data. The shift and scale parameters provide efficient and interpretable data summaries, while the common shape function is usually modeled nonparametrically, to provide flexibility. However, due to the deterministic nature of the shift and scale parameters, potential correlations within a subject are ignored. Furthermore, the shape of the common function may depend on other factors, such as disease. In this article, we propose shape-invariant mixed effects models. A second-stage model with fixed and random effects is used to model individual shift and scale parameters. A second-stage smoothing spline ANOVA model is used to study potential covariate effects on the common shape function. We apply our methods to a real data set to investigate disease effects on circadian rhythms of cortisol, a hormone that is affected by stress. We find that patients with Cushing's syndrome lost circadian rhythms and their 24-hour means were elevated to very high levels. Patients with major depression had the same circadian shape and phases as normal subjects. However, their 24-hour mean levels were elevated and amplitudes were dampened for some patients.