Mortality in various types of osteogenesis imperfecta.

BACKGROUND: Osteogenesis imperfecta (OI) is a group of closely related inherited diseases characterized by abnormal bone fragility. The current clinical classification system delineates 6 types, one of which (type II) is so severe that mortality is 100%, either intrauterine or perinatal. The types are differentiated by clinical groups, by severity, and by the presence or absence of other features such as blue sclerae or dentinogenesis imperfecta. There are no known previous studies of mortality in OI. RESULTS: From a registry created in association with the Brittle Bone Society, 743 patients with OI in England and Wales were observed in the period 1980-1993. These were classified into 3 groups (type IA, type III, and types IB, IVA, and IVB combined). Average annual mortality rates were determined in each group by sex and attained age. These rates were compared with 1981 rates in the population of England and Wales, matched by sex and age. Results are given in terms of exposures, observed and expected deaths, and 2 indices of excess mortality: mortality ratios and excess death rates per 1000 person-years. CONCLUSION: In type IA, 51.5% of the OI cases overall, there was no significant excess mortality (mortality ratio 108%, based on 15 deaths). In type III, on the other hand, excess mortality was very high in children and still significantly high at ages 15-34 years. In the combined group of types IB, IVA, and IVB, the mortality ratio was 157% in patients aged 45 and up (not significant at the 95% confidence level), but higher ratios at younger ages were significant, even though based on a total of only 5 deaths.

Long-term comparative mortality in hyperthyroid patients treated with radio-iodine, a cohort study in England.

BACKGROUND: The focus of this long-term study (entry 1950-1989, follow-up to 1996) on hyperthyroid patients treated with 131I was on all-cause mortality and some specific causes other than cancer. The study was carried out on 7209 patients who were residents of the Midlands area around Birmingham, England. In a total exposure of 105,028 patient-years, 3611 deaths were observed, compared with 3186 deaths expected from age/sex-matched rates in the English life tables. RESULTS: Data were reported for observed and expected deaths, and SMR values (standardized mortality ratios to one decimal place) by selected causes of death in combinations with duration, age, and 3 dosage groups of 131I. Exposure data were given only for the dosage groups of 131I, not for the results by age or duration. CONCLUSIONS: Comparative mortality was significantly increased for all-cause mortality (but with an SMR of only 1.13), and also for cardiovascular and cerebrovascular deaths, and deaths from fracture. Excess mortality was greatest in the first year after entry, and decreased thereafter. There was no reported increase in cancer mortality.

Long-term comparative cancer mortality after use of radio-iodine in the treatment of hyperthyroidism, a fully reported multicenter study.

BACKGROUND: The Cooperative Thyrotoxicosis Therapy Follow-up Study reported cancer mortality after use of radio-iodine (131I), surgery, and other treatments in patients with hyperthyroidism, started during 1946-1964 and followed through 1968. In 1984, the National Cancer Institute initiated a continued follow-up (FU) of the same cohort through 1990, to assess potential carcinogenic effects of 131I through 1990, a potential FU ranging from 26 to 44 years. RESULTS: With an initial cohort of 35,630 hyperthyroid patients, a mean entry age of 46 years and 738,831 patient-years of FU, the 2950 observed cancer deaths were only minimally over the 2858 expected cancer deaths from US population rates, after exclusion of first-year deaths. In the 64.7% of patients receiving 131I, the total cancer deaths were significantly above expected (258 versus 208), only at durations of 1-5 years, not thereafter. Thyroid cancer deaths were significantly elevated, but the number was very small, only 24 deaths in all exposures, as compared with 5.89 expected. The overall thyroid cancer mortality was higher in cases of toxic nodular goiter than in patients with Graves' disease. CONCLUSIONS: Despite the small early increase in total cancer deaths and the relatively higher increase in thyroid cancer deaths (only 18 excess deaths), the authors conclude that therapy with the doses of 131I used is a safe procedure. The interpretation of cause-specific excess mortality is greatly influenced by the prevalence of the cause of death.

Incidence of seizures after traumatic brain injury--a 50-year population survey.

BACKGROUND: Records of the Rochester Epidemiological Project were used to determine the incidence of secondary seizures after traumatic brain injury (TBI) in all cases treated for this condition in the population of Olmsted County, Minn, from 1935 to 1984. Medical records of the Mayo Clinic and all other medical facilities in Olmsted County, Minn, are in the database of this Project. Incidence rates after TBI were compared with incidence rates of idiopathic epilepsy previously determined for Olmsted County. RESULTS: TBI cases were divided into 3 defined severity categories: mild, moderate, and severe. Out of 4541 cases of TBI accumulated in 50 years only 97 cases developed 1 or more seizures (46 cases of seizure secondary to other definite causes were excluded). Incidence rates were highest in the first year after the head injury. The overall excess incidence rate was very low in mild TBI, only 0.3 per 1000 per year, but was higher in severe TBI, with an excess rate of 10 per 1000 per year. Only 7.2% of the TBI cases were classified as severe (loss of consciousness or amnesia for more than 24 hours, subdural hematoma, or brain contusion). CONCLUSION: The long-term incidence of seizures beyond the incidence rate of idiopathic epilepsy is low after mild or moderate TBI, but is at the rate of 10 excess cases per 1000 per year in the minority of cases with severe TBI.

The first mortality follow-up study: the 1841 Report of William Farr (physician) on the mortality of lunatics.

BACKGROUND: In the 1830s in England, there was a great cultural interest in the collection and publishing of all kinds of statistics. The Council of the Statistical Society of London (founded in 1834) commissioned one of its Fellows, Dr William Farr, to investigate and prepare a report on the mortality of patients in the county asylums, with the mortality in a large number of proprietary houses that were licensed to care for patients with mental illness (then called lunatics) ordered for confinement because of their mental condition. Committees of Parliament had investigated the condition of the mentally ill confined to the asylums and taken measures in an attempt to improve their treatment and to correct abuses. RESULTS: Farr collected data from Hanwell, the Middlesex County asylum opened in 1831, and other asylums on annual admissions, resident patients, deaths, discharges, derived years of residence (exposure to risk), and annual mortality rates by duration and as an aggregate. He used similar data from a recent report on a large number of licensed houses. For the best estimate of comparative mortality, an assumed age distribution by sex and rates from the English Life Table No. 1 (constructed by Farr for 1841). CONCLUSION: Farr demonstrated that annual mortality rates were higher at durations 0-1.5 years than at durations 1.5-7.5 years, higher in men than in women, higher in paupers than in other patients, higher in licensed houses than in the Hanwell Asylum, and higher in the 4 large licensed houses than in a collection of smaller ones. COMMENT: A brief sketch of Farr's life is given as a memorial tribute to his pioneer work in vital statistics, life table methodology, public health, and life insurance medicine.

Structured settlement annuities, part 1: overview and the underwriting process.

Structured settlement underwriting is the underwriting of medically impaired lives for the purchase of an annuity to fund the settlement. Other than risk assessment, structured settlement (SS) underwriting has little in common with traditional life insurance underwriting. Most noteworthy of these differences is the relative lack of actuarial data on which to base decisions about mortality and the necessity for prospective thinking about risk assessment. The purpose of this paper is to provide a foundation for understanding the structured settlement business and to contrast the underwriting of structured settlements with that of traditional life insurance. This is the first part of a two-part article on SS annuities. Part 2 deals with the mortality experience in SS annuitants and the life-table methodology used to calculate life expectancy for annuitants at increased mortality risk.

Structured settlement annuities, part 2: mortality experience 1967--95 and the estimation of life expectancy in the presence of excess mortality.

BACKGROUND: the mortality experience for structured settlement (SS) annuitants issued both standard (Std) and substandard (SStd) has been reported twice previously by the Society of Actuaries (SOA), but the 1995 mortality described here has not previously been published. We describe in detail the 1995 SS mortality, and we also discuss the methodology of calculating life expectancy (e), contrasting three different life-table models. RESULTS: With SOA permission, we present in four tables the unpublished results of its 1995 SS mortality experience by Std and SStd issue, sex, and a combination of 8 age and 6 duration groups. Overall results on mortality expected from the 1983a Individual Annuity Table showed a mortality ratio (MR) of about 140% for Std cases and about 650% for all SStd cases. Life expectancy in a group with excess mortality may be computed by either adding the decimal excess death rate (EDR) to q' for each year of attained age to age 109 or multiplying q' by the decimal MR for each year to age 109. An example is given for men age 60 with localized prostate cancer; annual EDRs from a large published cancer study are used at duration 0-24 years, and the last EDR is assumed constant to age 109. This value of e is compared with e from constant initial values of EDR or MR after the first year. Interrelations of age, sex, e, and EDR and MR are discussed and illustrated with tabular data. CONCLUSIONS: It is shown that a constant MR for life-table calculation of e consistently overestimates projected annual mortality at older attained ages and underestimates e. The EDR method, approved for reserve calculations, is also recommended for use in underwriting conversion tables.

Mortality in two recent reports of clinical trials on patients with congestive heart failure compared with mortality in three previous clinical trials.

BACKGROUND: Several clinical trials of drug treatment of patients with congestive heart failure (CHF) have previously been reported as Mortality Abstracts in the Journal of Insurance Medicine. Results are presented here for two similar clinical trials reported in September 1999 and compared with the previous results. RESULTS: In a recent international multicenter clinical trial, excess mortality in terms of excess death rates (EDRs) was reduced from 195 per 1000 per year in the placebo group to 139 in the group treated with Spironolactone. There was no significant reduction in the Danish multicenter study of Dofetilide to convert the atrial fibrillation (AF) to a normal rhythm in the 25% of the CHF patients who had AF (EDR was 224 in the placebo group and 216 in the Dofetilide group). In both of these studies, there were more patients with severe CHF than in the previous studies and the EDR values were higher. Results from the Danish study by severity according to the New York Heart Association (NYHA) classification show a progressive increase in EDR from 173 in class 2 to 237 in class 3 to 392 in class 4. CONCLUSION: Excess mortality in symptomatic CHF is far outside the issue limits for individual life insurance, but these results are of potential utility for the underwriting of such cases for structured settlement annuities.

Comparative mortality in cerebral palsy patients in California, 1980-1996.

BACKGROUND: The large database of the California Department of Developmental Services provides a data source for mortality rates in persons with mental retardation by age, sex, severity, cause and associated conditions. This study involves patients with a diagnosis of cerebral palsy. RESULTS: After a table of demographic data, four tables are used to show detailed age-related observed and expected mortality rates for Cerebral Palsy patients by sex and a severity factor that divides the patients into two groups of approximately equal size. The factor used was quadriplegia (all four limbs involved in motor dysfunction). Spasticity was the predominant feature of the motor dysfunction. CONCLUSION: Excess mortality was moderate in the less severe Cerebral Palsy patients, but was higher in those with quadriplegia (overall EDR--Excess Death Rate--about 6 per 1000 and 16 per 1000, respectively). In less severe cases EDR was higher at ages 1-4 years, the almost constant to age 49, then rose with advancing age. In case with quadriplegia EDR decreased in childhood and young adults to a relatively stable minimum at ages 25-49, then increased at older ages. There was little sex difference in EDR.

Mortality in rheumatoid arthritis: experience in four clinical series compared with the experience in the 1983 medical impairment study.

BACKGROUND: Survival curves and decimal Standardized Mortality Ratios (SMRs) were reported in 1994 for four clinical series of patients with Rheumatoid Arthritis (RA), with use of a common data processing system. It was felt to be desirable to compare the excess mortality observed in clinical patients with the excess mortality found in substandard policyholders with RA in the 1983 Medical Impairment Study. RESULTS: The weighted mean SMR was 2.25 for the male RA patients and 2.42 for the females. Although similar in magnitude to the percentage male and female Mortality Ratios (MRs) in the 1983 Medical Impairment Study, the derived Excess Death Rate (EDR) was higher for the clinical RA patients, despite the lower select mortality versus the population mortality, even after adjustment for difference in mean age between the clinical and insurance series. CONCLUSION: The higher EDR in four clinical series of RA patients probably reflects a higher proportion of more severe cases, who are declined for insurance or do not accept a highly rated offer.

How to construct a Mortality Abstract (number 497M-1 as an example).

BACKGROUND: Since the 1976 publication of the first Medical Risks volume of analyses of mortality follow-up (FU) studies (called "mortality abstracts"), members of the American Academy of Insurance Medicine have sought to extend this method of using FU studies in the medical literature to produce comparative mortality data useful for medical underwriting. Seminars on life table methodology started in 1977 and courses are now sponsored by the Board of insurance Medicine. The Journal of Insurance Medicine has published many mortality abstracts and methodology articles, and methodology is further pursued by the Editor in this issue of the Journal. TEXT: Retrieval of the source data (Annual Reports) is described and the source tables are evaluated. These are related to a previous 1976 Abstract. The design of the Abstract is described: other sources, tables, background, expected mortality, descriptive text, and comment. Aspects of life table methodology are used throughout the description. SUMMARY: The construction of the current Mortality Abstract 497M-1 on mortality in cystic fibrosis is described in detail.

Cystic fibrosis mortality: registry data of cystic fibrosis.

BACKGROUND: The Cystic Fibrosis Foundation has been collecting mortality and other data on cystic fibrosis patients from over 100 U.S. clinical centers since 1966, and publishes annual reports of the data. Cystic fibrosis is an important autosomal recessive genetic disease, affecting multiple body systems, with disability and early mortality as very common complications. RESULTS: Observed data by age in the 1994 and 1995 REports are compared with the earliest Registry experience. 1966-1972, after age-matching observed with expected mortality in the U.S. population. Observed and expected results are given in four tables of comparative mortality. CONCLUSION: Excess mortality as a rate difference has shown a marked decrease during this time period, ranging from 96% at age 0-1 to 52% at ages 20-25 years. This can be attributed to improvements in medical treatment.

Comparative mortality in mentally retarded patients in California, with and without Down's syndrome, 1986-1991.

BACKGROUND: The large database of the California Department of Developmental Services provides a data source for mortality rates in persons with mental retardation by age, sex, severity, cause and associated conditions. RESULTS: Data for 1986-1991 were used to determine age-related mortality rates in cases with and without Down's Syndrome, in three severity levels of mental retardation. Distribution data for the data-base population are also given. CONCLUSION: In both Down's and non-Down's patient groups excess mortality increased with severity of mental retardation at all ages. In both groups excess mortality tended to decrease with advancing age to age 35-39 years, and to increase at the older ages. The age increase in excess death rate at age 40 years and up was steeper in the patients with Down's Syndrome.

Randomized trial of carvedilol in treatment of congestive heart failure.

BACKGROUND: Members of the U.S. Cardevivol Study Group at Lovelace Scientific Resources, Albuquerque, NM, designed and carried out a large multicenter randomized clinical trial of Cardevilol vs placebo in patients with clinical heart failure and ejection fraction < 0.35. RESULTS: Derived exposure and death data at 50-day intervals to 400 days were used to construct tables of comparative mortality, all ages, both sexes and all durations to 400 days combined, in the Carvedilol and the placebo groups Derivation of expected mortality is explained in detail. CONCLUSIONS: Excess mortality, measured as mortality ratio or excess death rate, was much higher in the placebo than in the Cardevilol-treated group.

First-year acute coronary event rates in patients with acute myocardial infarction evaluated 5-7 days post-MI for ST depression in an ambulatory monitored ECG.

BACKGROUND: To present rates for fatal and non-fatal acute coronary events (ACE) in patients monitored 5-7 days after acute myocardial infarction, for abnormal ST depression in a continuous ambulatory ECG, with follow-up to the end of the first year after hospital admission. Comparative first-year mortality rates are also derived. RESULTS: The comparative mortality in the ischemic group is at an excess death rate of 92 per 1,000 per year as opposed to the nonischemic group where it was 15 per 1,000 per year. Nonfatal ACE rates were 346 and 154 per 1,000 per year, respectively, in the two groups. CONCLUSION: The risk of death is extremely high in the first few minutes and the first hour after the onset of an acute MI, and the risk diminishes steadily over the ensuing days weeks and months.