Defense of breast cancer malpractice claims.

The goal of this study was to determine whether factors associated with the successful defense and cost of malpractice cases involving the failure to diagnose breast cancer could be identified in medical and legal records. Secondary goals were to develop a multidisciplinary clinical algorithm utilizing National Comprehensive Cancer Network (NCCN) practice guidelines with practitioner risk management strategies. Physician deviations from these guidelines were tracked to identify high-risk areas in the diagnosis of breast cancer. A multidisciplinary clinical algorithm was introduced and practitioner risk management issues were addressed. In this study specific medical, legal, and cost factors were retrospectively abstracted and analyzed to identify associations between medical and legal factors and medicolegal outcome. ProMutual handled 156 malpractice cases involving breast cancer between January 22, 1986, and November 20, 1997. Of the total, 124 cases involving 212 defendants were closed. The closed cases were analyzed, using multivariable stepwise logistic and linear regression, to identify associations between clinical factors and case outcome. Women's health practitioners (WHPs), including obstetrician-gynecologists (OB-GYNs), family medicine, and internal medicine clinicians, were the largest group of defendants (97). Others included radiologists (43), surgeons (33), and pathologists (3). OB-GYNs accounted for 31% of these defendants, with a cost of more than $16 million. The greatest number of specialists represented in the open cases were radiologists, with 38% of the total. The defense model predicts that the probability of successful defense is lessened with inadequate record keeping, a patient that has metastasis and is alive, and a delay in diagnosis of 12 months or more. The overall indemnity model predicts a higher indemnity with the spread of disease at the time of evaluation, a patient who has metastasis and is alive, and a date of occurrence closer to the present. Indemnity is less in patients who have had a lymph node dissection, who have died, or who are alive without metastasis. The WHP model predicts an increased overall indemnity with the spread of disease at the time of evaluation and the presence of a mass without pain. Indemnity decreases with a history of pregnancy, absence of presenting symptoms, or presentation with pain with or without a mass, and the performance of a lymph node dissection.

Statistical models of outcome in malpractice lawsuits involving death or neurologically impaired infants.

The objective was to determine whether factors could be identified in medical and legal records that are associated with the successful defense of obstetrical malpractice cases involving the death or neurological impairment of infants. Obstetrical claims (169) closed by PROMUTUAL between January 1, 1990, and December 31, 1994, were retrospectively abstracted and analyzed to identify associations between medical and legal factors, and the medicolegal outcome. Multivariable analysis identifies that the use of pitocin, diagnosis of asphyxia, a delay in delivery, and the use of multiple defense expert witnesses decreased the chances of a successful defense. Two statistical models explaining indemnity payment were developed. The first, based on medical outcome, showed an increased indemnity payment when a case involved major neurological deficits, diagnosis of asphyxia, newborn seizures, later year of delivery, and participation of a particular defense firm. Perinatal or childhood death and the use of pitocin were indicators of a decrease in payment. The second model was based on long-term care requirements. In this model, indicators of increased indemnity payment were: nonreassuring intrapartum fetal heart rate tracing, later year of delivery, intensity of long-term care required, and participation of a particular defense law firm. Perinatal or childhood death, the use of pitocin, and settlement date increasingly removed from the occurrence date were the determinants of decreased payments in this model. Finally, the presence of major neurological deficits, the prolongation of a case, and the involvement of multiple law firms and defense witnesses increased the expense charged to and paid by the insurance company. Using the medical, legal, and financial data relevant to 169 obstetrical cases closed by one malpractice insurance carrier between 1990 and 1994, statistical models with potential predictive values for future malpractice claims involving neurologically impaired infants were constructed. These models may help determine in advance the chance a future case has for successful defense and the likely amount of expense and indemnity dollars that will be paid out to settle and defend it.

The effect of changes in atmospheric pressure on the occurrence of the spontaneous onset of labor in term pregnancies.

OBJECTIVE: Our purpose was to determine whether there is a relationship between changes in atmospheric pressure and spontaneous onset of labor in term pregnancy. STUDY DESIGN: All women admitted to Medical Center of Central Massachusetts-Memorial Hospital with spontaneous onset of labor at term and who were delivered on the service during a 12-month period represent the cohort for this study. Each maternal chart was abstracted to ensure that each member of the cohort met the inclusion criteria. Hourly recordings of atmospheric pressure made at the Worcester Station of the National Weather Service, Department of Commerce, were used as the meteorologic data points of interest. Least-squares regression was used to determine an equation that expresses the probability of the onset of labor in this cohort as a function of gestational age, which was used to calculate expected numbers for the statistical analyses. Two relationships were studied: (1) the ratio of the observed to the expected number of onsets of labor and (2) the initiation of labor and atmospheric pressure changes in the preceding 3 hours. RESULTS: Three-hour periods of falling atmospheric pressure were less often followed by initiation of labor than were the periods with other types of pressure sequences. No association was observed between the onset of labor and days of low mean pressure. CONCLUSION: Although there was an observed statistically significant association between falling barometric pressure and onset of labor, the magnitude of the difference is not of clinical significance.

A statistical model for predicting the outcome in breast cancer malpractice lawsuits.

OBJECTIVE: To analyze specific medical, legal, and cost factors that predict the probability of successfully defending lawsuits filed because of failure to diagnose breast cancer. METHODS: Seventy-six malpractice cases handled by the Massachusetts Medical Professional Insurance Association between June 29, 1983 and December 30, 1993 were abstracted and analyzed using univariate analysis, multivariate stepwise logistic and least-square regression analysis, and the Cox proportional hazards model to identify statistically significant associations between clinical factors and medicolegal outcomes. RESULTS: Obstetrician-gynecologists were defendants in the largest number of cases (38) and incurred the highest total indemnity ($7,629,570). The probability of defending a suit successfully increased with smaller tumor size and younger patients (less than 40 years of age). The failure to perform a biopsy was associated with a decreased probability of successful defense. Variables predicting high case cost included younger patient age, an increased length of delay in diagnosis, and the failure to perform a biopsy. The presence of metastasis at diagnosis was associated with an increased interval from diagnosis to the initiation of a suit. CONCLUSION: Statistical models that use medicolegal and cost factors can predict both the probability of a successful defense and the total cost of a breast cancer malpractice case.

Factors predictive of seizures among intensive care unit patients with gram-negative infections.

From the medical records of 238 intensive care unit (ICU) patients who had infections with gram-negative pathogens commonly associated with serious illness, we developed a predictive score of clinical risk factors for seizures. To evaluate the predictive ability of this score, we applied it to a separate population of 645 seriously ill hospitalized patients with similar gram-negative infections who were in antibiotic clinical trials. The patients at highest risk were classified into one of the following three categories: (a) patients with major central nervous system (CNS) insults (CNS surgery, hemorrhage, infection, or other lesion within 1 month before hospital admission or any history of CNS neoplasia), (b) patients with a predisposing factor (renal impairment or a history of seizures) plus a precipitating factor (anoxic encephalopathy/coma or an acute hypotensive episode), and (c) patients with both renal impairment and a history of seizures. Receiver operating characteristic (ROC) curves were calculated in each of the two populations. The area under the ROC curve (AUC) represents the probability that the score would rank a randomly chosen patient who subsequently had a seizure as having had a greater prior level of seizure risk than a randomly chosen patient who did not experience a seizure. The AUC was 0.87 (SE = 0.05) for the original population used to develop the score and 0.81 (SE = 0.04) for the population used for the validation study. The clinical risk score, based on readily available information, provides a useful means to identify among seriously ill infectious disease service patients, those who are at highest risk for seizures. It also serves as a baseline for evaluating the non-drug-related risk factors for seizures in patients treated with antibiotics.

Congenital malformations among offspring exposed in utero to progestins, Olmsted County, Minnesota, 1936-1974.

Comparison of a cohort of 988 offspring exposed in utero to exogenous progestins with a matched cohort of unexposed offspring did not result in detection of an association of congenital anomalies with exposure. The conclusions are based primarily on outcomes of pregnancy with exposure to progesterone and 17 alpha-hydroxyprogesterone caproate, and may not apply to androgenic progestins. Offspring exposed to combinations of progestins and estrogens were excluded from this study and may have a different distribution of anomalies.

Males exposed in utero to diethylstilbestrol.

An increased frequency of various genitourinary anomalies, infertility, and testicular cancer among males has been reported to follow intrauterine exposure to diethylstilbestrol, but not all studies have confirmed an association. This study was designed to determine whether a cohort of males exposed in utero to diethylstilbestrol had a higher frequency of urogenital abnormalities than an unexposed cohort. Biases in selection of exposed and control participants were minimized. Of 828 exposed and 676 control men studied by medical-record review, 265 exposed men and 274 controls also underwent a special clinical examination. Overall, the data suggest that diethylstilbestrol exposure of males in utero did not increase their risk of genitourinary abnormalities, infertility, or testicular cancer. Previously reported increased frequencies of these abnormalities in diethylstilbestrol-exposed men may have resulted from selection biases or differences in diethylstilbestrol use, or both.

Cholecystectomy and gastric ulcer--an etiologic relationship?

To determine whether cholecystectomy might be a factor predisposing to gastric ulceration, we evaluated the frequency of prior cholecystectomy in the 181 residents of Rochester, Minnesota, who had gastric ulcer diagnosed at the Mayo Clinic during the 10-year-period from 1966 to 1975. This frequency was similar to that in a control population from the same city, matched for age, sex, and time of registration at the Mayo Clinic. This was also true for subgroups of benign gastric ulcer. However, the frequency of prior cholecystectomy was higher in patients with type 1 ulcer than in those with type 2 or 3 ulcer, probably as a result of the relative predominance of older women in the type 1 ulcer group. These data do not support the hypothesis that cholecystectomy may be harmful in predisposing to gastric ulceration.

Malignant melanoma in the resident population of Rochester, Minnesota.

The incidence of malignant melanoma in the population of Rochester, Minnesota, was studied through use of the countywide diagnostic indexing system at the Mayo Clinic. The average annual incidence of cutaneous melanoma (4.0/100,000 creude, 4.2 adjusted to the age structure of the US 1950 population) was similar to that reported by the Third National Cancer Survey for 1969 through 1971 for locations in the United States at the approximate latitude of Rochester. Unlike reports from other studies, no change in incidence rate during the 25 years 1950 through 1974 was detjected in Rochester. The crude annual incidence of malignant melanoma of the eye, also determined for the same period, was 1.3/100,000.

Paternal age, stillbirths and mutation.

Previous suggestions that accumulation of mutations in the germ line of ageing fathers causes an increased stillbirth rate were based on analyses of data which were heterogeneous for social variables whose effects were confounded with possible paternal age effects. This study was confined to the analysis of stillbirth rates of groups of women selected to be homogeneous for education, previous pregnancy outcomes, age, race and marital status. It is concluded that stillbirth rates do not increase with father's age independently of maternal variables. Neither accumulation of mutations in the paternal germ line nor other biological change associated with father's age can be inferred to cause an increase in risk of stillbirth with increasing paternal age.

Comparison of longitudinal and cross-sectional analysis: maternal age and stillbirth ratio.

Previous studies have reported a U or J shaped association between stillbirth ratio and maternal age and have led to attempts to determine the causes behind the association. These analyses have generally been based on cross-sectional data. This study contrasts the descriptions of the association resulting from cross-sectional and the more appropriate longitudinal analysis of the same body of data. It is concluded that cross-sectional analysis substantially underestimates the maternal age of which stillbirth ratio reaches a minimum for each birth order examined. Additionally, the basic shape of the relationship is sometimes altered. Weighted logistic polynomial regression is used to describe the associations. Traditional interpretations of the results of a cross-sectional analysis have been that women who postpone childbearing until the late twenties thereby increase their risk of stillbirth. A consequence of the present findings is the rejection of the claim that women who choose employment during the period of their early twenties automatically incur elevated risk of stillbirth in their postponed pregnancies.

Pregnancy wastage and age of mother among the Amish.

PIP: Reports of approximately 7500 pregnancies in reproductive histories collected by Colette Wiffler through personal interviews with Old Order Amish families of Illinois, Iowa, Missouri, and Wisconsin during the 1968 through 1973 period were analyzed to test a prediction: a society in which healthy women generally want large numbers of children and do not marry unusually uoung should exhibit a slower rate of increase in fetal death ratios with age of mother than the general US population. In this study, fetal deaths occurring after 7 months of gestation were called stillbirths; those occurring between 6 weeks and 7 months were termed miscarriages. Neonatal deaths occurred within the 1st week following live birth. All loss ratios were calculated as the number of the specified type of pregnancy loss/1000 pregnancies which lasted at least 7 months. The minimum miscarriage and stillbirth ratios each occurred in the early 30s, but the ratios were not statistically different from those for mothers in their early 20s. The interpretation of the observation is complicated by substantial reductions in pregnancy wastage experienced by the general population over the long span of time (1898 through 1972) covered by the present data. For the US both late fetal death ratios and neonatal death rates specific for the age of the mother reach their minimum in the early 20s. While most available data provide information about late fetal death only, the study of pregnancies in New York's Health Insurance Plan revealed markedly higher fetal death ratios for mothers in the early 30s than for mothers in their 20s both for gestations of 12-19 weeks and for those of less than 12 weeks. Thus, the Amish fetal deaths differ from the general US pattern similarly for miscarriages and for the less numerous stillbirths. These results are compatible with the prediction under test but conflict with the expectations of the traditional idea that women in their early 20s have their ability to carry pregnancies to live birth impaired by age. The findings suggest that any increase in risk of fetal death caused by increasing age of an individual mother must be unimportant before age 35. It appears that women who decide to postpone their pregnancies until their late 20s or early 30s are probably not materially increasing the risk of fetal death. The same appears to be the case for early infant mortality.^ieng