Claims Experience Research Articles

Hierarchical Insurance Claims Modeling

This work describes statistical modeling of detailed, microlevel automobile insurance records. We consider 1993–2001 data from a major insurance company in Singapore. By detailed microlevel records, we mean experience at the individual vehicle level, including vehicle and driver characteristics, insurance coverage, and claims experience, by year. The claims experience consists of detailed information on the type of insurance claim, such as whether the claim is due to injury to a third party, property damage to a third party, or claims for damage to the insured, as well as the corresponding claim amount. We propose a hierarchical model for three components, corresponding to the frequency, type, and severity of claims. The first model is a negative binomial regression model for assessing claim frequency. The driver’s gender, age, and no claims discount, as well as vehicle age and type, turn out to be important variables for predicting the event of a claim. The second is a multinomial logit model to predict the type of insurance claim, whether it is third-party injury, third-party property damage, insured’s own damage or some combination. Year, vehicle age, and vehicle type turn out to be important predictors for this component. Our third model is for the severity component. Here we use a generalized beta of the second kind of long-tailed distribution for claim amounts and also incorporate predictor variables. Year, vehicle age, and person’s age turn out to be important predictors for this component. Not surprisingly, we show a significant dependence among the different claim types; we use a t-copula to account for this dependence. The three-component model provides justification for assessing the importance of a rating variable. When taken together, the integrated model allows more efficient prediction of automobile claims compared with than traditional methods. Using simulation, we demonstrate this by developing predictive distributions and calculating premiums under alternative coverage limitations.

Medical Malpractice and the Chest Physician

The US malpractice system is based on tort law, which holds physicians responsible for not harming patients intentionally or through negligence. Malpractice claims are brought against physicians from most medical disciplines in proportion to their numbers in practice and to the frequency with which they perform procedures. Claims against chest physicians most commonly allege injuries caused by the following: (1) errors in diagnosis, (2) improper performance of procedures, (3) failure to supervise or monitor care, (4) medication errors, and (5) failure to recognize the complications of treatment. Most of these injuries occur in hospitals, and many of the injured patients die. The social goals of the medical malpractice system include the following: (1) compensating patients injured through negligence, (2) exacting corrective justice, and (3) deterring unsafe practices by creating an economic incentive to take greater precautions. Some patients injured through negligence are compensated, but most are not. Claims are brought against some negligent physicians but also some who are not negligent, and being negligent does not guarantee that a claim will be brought. The deterrent effect of medical malpractice is unproven, and the malpractice system may prompt defensive medicine and increase health-care costs. And by stressing individual accountability, it conflicts with a systems-oriented approach to reducing medical errors.

Suits And Suitability

Suits And Suitability

The Impact of Malpractice Burden on Michigan Obstetrician-Gynecologists' Career Satisfaction

Medical services for pregnancy and childbirth are inherently risky and unpredictable. In many states, obstetrician-gynecologists (OB-GYNS) who attend the majority of childbirths in the United States and provide the most clinically complex obstetric procedures are struggling with increasing malpractice insurance premiums and litigation risk. Despite its significant implications for patient care, the potential impact of malpractice burden on OB-GYN physicians' career satisfaction has not been rigorously tested in previous research. Drawing on data from a statewide survey of obstetric providers in Michigan, this paper examined the association between medical liability burden and OB-GYNs' career satisfaction. Malpractice insurance premiums and malpractice claims experience were used as 2 objective measures for medical liability burden. Descriptive statistics were calculated and multivariable logistic regressions estimated for data analysis. Although most respondents reported satisfaction with their overall career in medicine, 43.7% had become less satisfied over the last 5 years and 34.0% would not recommend obstetrics/gynecology to students seeking career advice. Multivariable regression analysis showed that compared to coverage through an employer, paying > or =$50,000/year for liability insurance premium was associated with lower career satisfaction among OB-GYNs (odds ratio, 0.35; 95% confidence interval, 0.13-0.93). We found no significant impact of malpractice claims experience, including both recent malpractice claims (during the last 5 years [2001--2006]) and earlier malpractice claims (>5 years ago), on overall career satisfaction. The findings of this study suggest that high malpractice premiums negatively affect OB-GYN physicians' career satisfaction. The impact of the current medical liability climate on quality of care for pregnant women warrants further investigation.

Medical Malpractice in Gastroenterology

Gastroenterologists commonly perceive themselves to be at increased legal risk because they perform invasive procedures. However, there is little published information about gastroenterology (GI) malpractice claims. The goal of this study was to evaluate available malpractice claim data within GI. This study was a database analysis of GI claims submitted by insurance companies to the Physician Insurers Association of America Data Sharing Project from January 1, 1985, to December 31, 2005. Another analysis from 2005 compared GI claims with other subspecialties. Sixty-six percent of physicians involved in claims had previous claims experience. The most common reasons for claims were errors in diagnosis (28%) and improper performance of a procedure (25%). Seventy-two percent of reported closed claims were settled out of court. Of 12,367 total claims in 2005, only 233 (1.8%) were within GI. GI ranked below other procedurally based subspecialties in numbers of claims per physician. GI does not rank highly among subspecialties in malpractice claims and only a minority of claims are procedure-related. Physicians with claims experience are likely to have further claims against them and should consider evaluating their practices.

Malpractice Risk: Trauma Care Versus Other Surgical and Medical Specialties

Medical malpractice has been noted to play an important role in physicians' decisions to pursue or remain in certain presumed high-risk specialties such as trauma surgery, despite little evidence suggesting an elevated malpractice risk. The objective of this study was to compare the malpractice experience for trauma care and other medical and surgical specialties at an academic medical center. Information regarding all potentially compensable medical events (hereafter "events") and actual lawsuits that occurred between 2003 and 2006 at one academic medical institution, including the department or service primarily involved, the current medical-legal disposition of the event, and the actual or expected expenses was obtained. The number of patients admitted to each service and the time they spent in the hospital was also obtained, and the number of events per capita and length of stay was calculated. Among the 13 medical and surgical specialties considered, there were 194 total events, 183,392 patients seen, and 757,880 days of hospitalization. The trauma service had the fewest events and lawsuits per 10,000 patient-days and ranked 10th (11th for lawsuits) on a per capita basis, and 9th in total estimated cost. With the fewest events and lawsuits per patient days and a relatively low amount set aside for claims, trauma care has better claim experience than almost every other clinical service line. Although these results have some limitations, they refute the perception that trauma care is a higher medicolegal risk. This observation should not be cited as a disincentive for surgeons to provide trauma care.

Deterministic claims reserving in short-term insuarance contracts

Claims reserving for general insurance business has developed significantly over the recent past. This has been occasioned by the growth of the insurance market, with the risk underwriting process becoming more and more complex. New insurance products have been developed that cater for the more specific needs of the policyholder. Latent claims have also arisen in recent years, putting major strains on company resources. The case of asbestosis related claims testifies to this, having received widespread attention. Furthermore, recent disasters, such as the floods in Europe and the September 11th terrorist attacks on the U.S. have contributed to the need for more complex ways of analyzing claims experience. The suitability of the models used in claims reserving, have had to be reviewed to ensure that they do not give false impressions. The object of this paper, therefore, is to come up with a comparison of different methods of claims' reserving for a general insurer with a given claims' experience. The suitability of each of the estimates is noted to depend on the purpose of the reserving exercise. The paper discusses some of the methods (for instance, the basic chain ladder method, inflation adjusted chain ladder method, separation technique and Bornhuetter-Fergusson technique) used in claims' reserving, and for a particular claims experience, it gives an analysis of how well each of the methods models claims experience. > East African Journal of Statistics Vol. 1 (2) 2006: pp. 198-214

Nearest-neighbour methods for reserving with respect to individual losses

This paper focuses on the problem of predicting individual future late claims within a heterogeneous portfolio. For this purpose, thek-nearest-neighbour method is analysed, which intends to carry forward a considered loss according tok appropriately weighted similar observed claims experiences. Using a nonparametric regression approach, a normal approximation and asymptotic confidence intervals for the expected future indemnities are derived. The application of a backtesting procedure to third party liability insurance data reveals that thek-nearest-neighbour estimator performs better than the Chain-Ladder method.

Managing NHS records: complying with the new code of practice

PurposeThe purpose of this article is to assess the new NHS code of practice on records management, published in March 2006.Design/methodology/approachThe guidelines are put into the context of good practice, and changes to the way that medical practice is being delivered. Recent MPS experience of complaints and claims is also used to inform the analysis.FindingsThe code of practice includes much that is useful, in particular reviewing the legislation on the subject and providing more detailed retention schedules than the previous guidance. However, there are also some omissions, and some areas where more detail would have been helpful.Practical implicationsHealth professionals can take advantage of the new code of practice to refresh their knowledge and, if they have responsibility, review the records management within their own organisation.Originality/valueThe paper highlights the continuing risks of substandard records management.

Ruin probabilities and aggregrate claims distributions for shot noise Cox processes

We consider a risk process R t where the claim arrival process is a superposition of a homogeneous Poisson process and a Cox process with a Poisson shot noise intensity process, capturing the effect of sudden increases of the claim intensity due to external events. The distribution of the aggregate claim size is investigated under these assumptions. For both light-tailed and heavy-tailed claim size distributions, asymptotic estimates for infinite-time and finite-time ruin probabilities are derived. Moreover, we discuss an extension of the model to an adaptive premium rule that is dynamically adjusted according to past claims experience.

Inappropriate Drug Combinations Among Privately Insured Patients With HIV Disease

The objective of this study was to examine inappropriate drug combinations among privately insured patients with HIV disease. Data were obtained from the MarketScan Commercial Claims and Encounter Database for the years 1999 and 2000. Each of the 2110 person-years of data examined in this study represents the claims experience of an enrollee with HIV disease who filled an antiretroviral medication prescription in either 1999 or 2000 for a protease inhibitor or nonnucleoside reverse transcription inhibitor. This study compares the claims experience of patients with HIV who filled a prescription for an inappropriate drug combination as specified in guidelines jointly issued by the U.S. Department of Health and Human Services and the Henry J. Kaiser Family Foundation with the claims experience of patients who did not. An inappropriate drug combination was found in approximately 2% of the person-years of data, and persons who experienced an inappropriate drug combination had higher claims costs. One half of all of the inappropriate drug combinations involved a single lipid-lowering agent (simvastatin). Protease inhibitors decrease the activity of the enzyme that metabolizes simvastatin, and high concentrations of simvastatin have been associated with muscle damage. We found that patients who received protease inhibitors and simvastatin were more likely to experience muscle damage. Persons with HIV have compromised immune systems and often take many medications. Thus, the risk and consequences of medication errors are severe, and both providers and patients should carefully monitor drug regimens to ensure that they are both safe and efficacious.

An Accurate Asymptotic Approximation for Experience Rated Premiums

In the Bayesian approach, the experience rated premium is the value which minimizes an expected loss with respect to a posterior distribution. The posterior distribution is conditioned on the claim experience of the risk insured, represented by a n-tuple of observations. An exact analytical calculation for the experience rated premium is possible under restrictive circumstances only, regarding the prior distribution, the likelihood function, and the loss function. In this article we provide an analytical asymptotic approximation as n → ∞ for the experience rated premium. This approximation can be obtained under more general circumstances, it is simple to compute, and it inherits the good accuracy of the Laplace approximation on which it is based. In contrast with numerical methods, this approximation allows for analytical interpretations. When exact calculations are possible, some analytical comparisons confirm the good accuracy of this approximation, which can even lead to the exact experience rated premium.

An Accurate Asymptotic Approximation for Experience Rated Premiums

In the Bayesian approach, the experience rated premium is the value which minimizes an expected loss with respect to a posterior distribution. The posterior distribution is conditioned on the claim experience of the risk insured, represented by a n-tuple of observations. An exact analytical calculation for the experience rated premium is possible under restrictive circumstances only, regarding the prior distribution, the likelihood function, and the loss function. In this article we provide an analytical asymptotic approximation as n → ∞ for the experience rated premium. This approximation can be obtained under more general circumstances, it is simple to compute, and it inherits the good accuracy of the Laplace approximation on which it is based. In contrast with numerical methods, this approximation allows for analytical interpretations. When exact calculations are possible, some analytical comparisons confirm the good accuracy of this approximation, which can even lead to the exact experience rated premium.

Illness experience before and after an initiative to identify, treat, and prevent Helicobacter pylori-related diseases in the workplace.

Objective outcome measures are needed to confirm the merits of large-scale health initiatives in the workplace. We used aggregate medical claims data to evaluate the illness experience of 5160 employees during the 2 years after versus the 2 years before an intervention intended to identify, treat, and prevent Helicobacter pylori-related illnesses. Across all participants, a 2.1-fold reduction (95% confidence interval 1.4-3.1) in ulcer-related illness episodes and a marginal 1.1-fold reduction (95% confidence interval 0.9-1.4) in episodes due to other stomach and duodenal diseases were achieved. Improvement in claims experience was most notable among 250 employees with ulcer findings on the screening examination. This H. pylori health initiative appears to have contributed to a cost-effective reduction in subsequent illness episodes due to ulcer and possibly due to other stomach and duodenal disease.

Modelling repeated insurance claim frequency data using the generalized linear mixed model

Most of the methods used to estimate claim frequency rates in general insurance have assumed that data are independent. However, it is not uncommon for information stored in the database of an insurance company to contain previous years' claim data from each policyholder. We consider the application of the generalized linear mixed model approach to the analysis of repeated insurance claim frequency data in which a conditionally fixed random effect vector is incorporated explicitly into the linear predictor to model the inherent correlation. A motor insurance data set is used as the basis for simulation to demonstrate the advantages of the method. Ignoring the underlying association for observations within the same policyholder results in an underestimation of the standard error of the parameter estimates and a remarkable reduction in the prediction accuracy. The method provides a viable alternative for incorporating repeated claim experience that enables the revision of rates in general insurance.

Reduced medicolegal risk by compliance with obstetric clinical pathways: a case–control study

Objective To estimate whether guideline compliance affected medicolegal risk in obstetrics and whether malpractice claims data can provide useful information on guideline noncompliance by focusing on the claims experience of a large health system delivering approximately 12,000 infants annually. Methods We retrospectively identified 290 delivery-related (diagnosis-related groups 370–374) malpractice claims and 262 control deliveries at the health system during the period from 1988 to 1998. Clinical pathways for vaginal and cesarean delivery implemented in 1998 were used as a “standard of care.” We compared rates of noncompliance with the pathways in the claims and control groups, calculated an odds ratio for increased risk of being sued given departure from the guideline standards, and calculated the elevated risk of litigation introduced by noncompliance. We also compared the frequencies of different types of departures across claims and control groups. Results Claims closely resembled controls on several descriptive measures (mother’s age, location of delivery, type of delivery, and complication rates), but noncompliance with the clinical pathway was significantly more common among claims than controls (43.2% versus 11.7%, P < .001; odds ratio = 5.76, 95% confidence interval 3.59, 9.2). In 81 (79.4%) of the claims involving noncompliance with the pathway, the main allegation in the claim related directly to the departure from the pathway. The excess malpractice risk attributable to noncompliance explained approximately one third (104 of 290) of the claims filed (attributable risk = 82.6%). There were no significant differences in the types of deviation from the guidelines across claims and control groups. Conclusion In addition to reducing clinical variation and improving clinical quality of care, adherence to clinical pathways might protect clinicians and institutions against malpractice litigation. Malpractice data might also be a useful resource in understanding breakdowns in processes of care.

Multivariate Credibility for Aggregate Loss Models

Credibility is a form of insurance pricing that is widely used, particularly in North America. It is a special type of experience rating that employs a weighted average of claims experience and a previously established price to determine a new price for each risk class under consideration. This article extends traditional credibility formulas in two aspects. The new procedures are called “multivariate credibility” because both aspects make use of additional sources of data when compared to traditional formulas. Specifically, the first portion of the paper considers data from both the claims number and claims amount processes. Assuming an aggregate loss model for total claims, optimal insurance pricing formulas are derived. The insurance prices turn out to be an intuitively appealing weighted average of the overall mean claim, the claims number experience, and the claims amount experience. The second portion of the paper considers data from claims number and amount processes from multiple lines of business. By using covariances among lines of business (that are conditional on the unobserved heterogeneity), this article shows how to derive more efficient insurance prices. Accounting for covariance among different random quantities (securities) is standard practice in the investment industry. It is more difficult in an insurance context because of the heterogeneity associated with different risk classes. Nonetheless, ignoring this covariance has important ramifications, both theoretically and practically. For an illustrative sample of Massachusetts automobile claims, we show that the relative differences in accounting for and ignoring the covariance range from −3.9% to 14.5% for a selected bundle of insurance coverages.

Disabling occupational injury in the US construction industry, 1996.

In 1996 the US construction industry comprised 5.4% of the annual US employment but accounted for 7.8% of nonfatal occupational injuries and illness and 9.7% of cases involving at least a day away from work. Information in the published literature on the disability arising from construction injuries is limited. The construction claims experience (n = 35,790) of a large workers' compensation insurer with national coverage was examined. The leading types and sources of disabling occupational morbidity in 1996 in the US construction industry were identified. Disability duration was calculated from indemnity payments data using previously published methods. The average disability duration for an injured construction worker was 46 days with a median of 0 days. The most frequently occurring conditions were low back pain (14.8%), foreign body eye injuries (8.5%), and finger lacerations (4.8%). Back pain also accounted for the greatest percentage of construction claim costs (21.3%) and disability days (25.5%). However, the conditions with the longest disability durations were sudden-onset injuries, including fractures of the ankle (median = 55 days), foot (42 days), and wrist (38 days). Same-level and elevated falls were the principal exposures for fractures of the wrist and ankle, whereas elevated falls and struck by incidents accounted for the majority of foot fractures. Manual materials handling activities were most often associated with low back pain disability. The results suggest that these most disabling injuries can be addressed by increasing primary prevention resources in slips and falls and exposures related to injuries of sudden-onset as well as in reducing manual materials handling and other exposures associated with more gradual-onset injuries.

Defensive practice among psychiatrists: a questionnaire survey

Objective: There has been little research on the prevalence of defensive practice within hospital settings. The aim of this report was to examine the extent of defensiveness among psychiatrists and...

The combined effect of delay and feedback on the insurance pricing process: a control theory approach

Further to De Finetti’s [Su una impostazione alternativa della theoria collectiva del rischio. In: Transactions of the 15th International Congress of Actuaries, Vol. II, New York, pp. 433–443] proposal, of a modified random walk for the (accumulated) surplus reserve ( S) of an insurance system with a reflecting barrier at a predefined level, we design a model similar to that of Balzer and Benjamin [J. Inst. Actuaries 107 (1980) 513], involving a smooth control action. Given the basic difference equation, which describes the development of the surplus process and the delays inherent to an insurance system, we propose a particular decision function for the determination of the premium ( P). For this purpose, we use the recent claim ( C) experience and a negative feedback mechanism based on the latest known surplus value. The model assumes that the delay factor ( f) is a free control parameter with a constant accumulation factor ( R) for the surplus reserve. We investigate the stability of the system and the optimal parameter design (in terms of the fastest response and return to the initial or steady state). We determine appropriate values for the feedback factor ( ε) under the specific premium decision function using the tools of control theory. One of the results is the derivation of a critical value for the delay factor ( f ∞) beyond which instability is certain irrespective of the choice of the feedback factor ( ε).