Medical Record Linkage Research Articles

Distinguishing patterns in the dynamics of long-term medication use by Markov analysis: beyond persistence.

BackgroundIn order to accurately distinguish gaps of varying length in drug treatment for chronic conditions from discontinuation without resuming therapy, short-term observation does not suffice. Thus, the use of inhalation corticosteroids (ICS) in the long-term, during a ten-year period is investigated. To describe medication use as a continuum, taking into account the timeliness and consistency of refilling, a Markov model is proposed.MethodsPatients, that filled at least one prescription in 1993, were selected from the PHARMO medical record linkage system (RLS) containing >95% prescription dispensings per patient originating from community pharmacy records of 6 medium-sized cities in the Netherlands.The probabilities of continuous use, the refilling of at least one ICS prescription in each year of follow-up, and medication free periods were assessed by Markov analysis. Stratified analysis according to new use was performed.ResultsThe transition probabilities of the refilling of at least one ICS prescription in the subsequent year of follow-up, were assessed for each year of follow-up and for the total study period.The change of transition probabilities in time was evaluated, e.g. the probability of continuing ICS use of starters in the first two years (51%) of follow-up increased to more than 70% in the following years. The probabilities of different patterns of medication use were assessed: continuous use (7.7%), cumulative medication gaps (1–8 years 69.1%) and discontinuing (23.2%) during ten-year follow-up for new users. New users had lower probability of continuous use (7.7%) and more variability in ICS refill patterns than previous users (56%).ConclusionIn addition to well-established methods in epidemiology to ascertain compliance and persistence, a Markov model could be useful to further specify the variety of possible patterns of medication use within the continuum of adherence. This Markov model describes variation in behaviour and patterns of ICS use and could also be useful to investigate continuous use of other drugs applied in chronic diseases.

Age- and Sex-Associated Trends in Bloodstream Infection

Despite increasing concerns about antimicrobial resistance and emerging pathogens among blood culture isolates, contemporary population-based data on the age- and sex-specific incidence of bloodstream infections (BSIs) are limited. Retrospective, population-based, cohort study of all residents of Olmsted County, Minnesota, with a BSI between January 1, 2003, and December 31, 2005. The medical record linkage system of the Rochester Epidemiology Project and microbiology records were used to identify incident cases. A total of 1051 unique patients with positive blood culture results were identified; 401 (38.2%) were classified as contaminated. Of 650 patients with cultures deemed clinically relevant, the mean +/- SD age was 63.1 +/- 23.1 years, and 52.5% were male. The most common organisms identified were Escherichia coli (in 163 patients with BSIs [25.1%]) and Staphylococcus aureus (in 108 patients with BSIs [16.6%]). Nosocomial BSIs were more common in males than females (23.8% vs 13.9%; P = .002). The age-adjusted incidence rate of BSI was 156 per 100 000 person-years for females and 237 per 100 000 person-years for males (P<.001), with an age- and sex-adjusted rate of 189 per 100 000 person-years. Rates of BSI due to gram-positive cocci were 64 per 100 000 person-years for females and 133 per 100 000 person-years for males (P<.001); gram-negative bacillus BSI rates (85/100 000 person-years for females and 79/100 000 person-years for males) were not significantly different between sexes (P = .79). The rate of S aureus BSI was 23 per 100 000 person-years for females and 46 per 100 000 person-years for males (P = .005). There are significant differences in the age and sex distribution of organisms among patients with BSIs. The incidence of BSI increases sharply with increasing age and is significantly higher in males, mainly because of nosocomial organisms, including S aureus.

Incidence and outcome of cervical artery dissection: A population-based study

Background: Incidence rates for internal carotid artery dissection (ICAD) have been reported to be 2.6 to 2.9 per 100,000, but reliable epidemiologic data for vertebral artery dissection (VAD) are not available. Objective: To determine the incidence rate of cervical artery dissection (CAD) in a defined population. Methods: With IRB approval, we used the medical record linkage system of the Rochester Epidemiology Project to identify all patients diagnosed with spontaneous ICAD and VAD for the period of 1987–2003 in Olmsted County, MN. Results: Of 48 patients with CAD, there were 32 patients with ICAD and 18 patients with VAD. In Olmsted County, the average annual incidence rate for ICAD was 1.72 per 100,000 population (95% CI, 1.13 to 2.32) and for VAD 0.97 per 100,000 population (95% CI, 0.52 to 1.4). The average annual incidence rate for CAD was 2.6 per 100,000 population (95% CI, 1.86 to 3.33). The most frequently encountered symptoms in CAD were head or neck pain (80%), cerebral ischemia (TIA or infarct) (56%), and Horner syndrome (25%). Good outcome (defined as modified Rankin score of 0 to 2) was seen in 92% of patients. No recurrence of dissection was observed during a mean 7.8 years of follow-up. Conclusions: Internal carotid artery dissection was detected approximately twice as frequently as vertebral artery dissection in the overall study, but in the latter half of the study period, vertebral artery and internal carotid artery dissection incidence rates were equivalent. The majority of cervical artery dissection patients in the community have excellent outcome, and contrary to many tertiary referral series, re-dissection is rare.

Incidence and outcome of cervical artery dissection

Incidence rates for internal carotid artery dissection (ICAD) have been reported to be 2.6 to 2.9 per 100,000, but reliable epidemiologic data for vertebral artery dissection (VAD) are not available. To determine the incidence rate of cervical artery dissection (CAD) in a defined population. With IRB approval, we used the medical record linkage system of the Rochester Epidemiology Project to identify all patients diagnosed with spontaneous ICAD and VAD for the period of 1987-2003 in Olmsted County, MN. Of 48 patients with CAD, there were 32 patients with ICAD and 18 patients with VAD. In Olmsted County, the average annual incidence rate for ICAD was 1.72 per 100,000 population (95% CI, 1.13 to 2.32) and for VAD 0.97 per 100,000 population (95% CI, 0.52 to 1.4). The average annual incidence rate for CAD was 2.6 per 100,000 population (95% CI, 1.86 to 3.33). The most frequently encountered symptoms in CAD were head or neck pain (80%), cerebral ischemia (TIA or infarct) (56%), and Horner syndrome (25%). Good outcome (defined as modified Rankin score of 0 to 2) was seen in 92% of patients. No recurrence of dissection was observed during a mean 7.8 years of follow-up. Internal carotid artery dissection was detected approximately twice as frequently as vertebral artery dissection in the overall study, but in the latter half of the study period, vertebral artery and internal carotid artery dissection incidence rates were equivalent. The majority of cervical artery dissection patients in the community have excellent outcome, and contrary to many tertiary referral series, re-dissection is rare.

Incidence of Primary Hyperparathyroidism in Rochester, Minnesota, 1993–2001: An Update on the Changing Epidemiology of the Disease

We updated the incidence of primary hyperparathyroidism in Rochester, Minnesota. The lower rates previously noted persisted, whereas parathyroidectomies at our institution remained high. These data suggest an etiologic factor may be responsible for the peak incidence in the 1970s. Automated serum calcium measurements were associated with a dramatic rise in primary hyperparathyroidism in the early 1970s, but a progressive decline in the incidence thereafter was unexpected and suggested a fundamental change in the epidemiology of the disease. Our objective was to evaluate trends in the incidence of primary hyperparathyroidism since 1992. In this population-based descriptive study, Rochester, MN, residents who met defined diagnostic criteria for primary hyperparathyroidism from January 1993 through December 2001 were identified through the medical record linkage system of the Rochester Epidemiology Project and the Mayo Clinic Laboratory Information System. Changes in incidence were evaluated by Poisson regression. Altogether, 136 Rochester residents (94 women and 42 men) were newly identified with primary hyperparathyroidism in 1993-2001. Their mean age was 56 years, and 93% had definite disease. The overall age- and sex-adjusted (to 2000 U.S. whites) rate during this period was 21.6 per 100,000 person-years, which was less than the annual rate of 29.1 per 100,000 observed in 1983-1992 and 82.5 per 100,000 in July 1974-1982. Although community incidence declined, the number of parathyroidectomies performed at our institution increased during the same period. Serum calcium was deleted from the automated chemistry panel in June 1996, but most subjects remained asymptomatic at diagnosis (95%) with mild hypercalcemia. The majority of subjects were observed without parathyroid surgery (75%), and there was minimal impact on patient management from the 1990 NIH consensus conference on asymptomatic primary hyperparathyroidism. The lower incidence of primary hyperparathyroidism noted through 1992 has persisted in our community through 2001, whereas parathyroidectomies at our institution remained high. These data suggest that some underlying etiologic factor, in addition to the introduction of automated serum calcium testing, may have been responsible for the peak incidence in the 1970s.

Medical record linkage in health information systems by approximate string matching and clustering.

BackgroundMultiplication of data sources within heterogeneous healthcare information systems always results in redundant information, split among multiple databases. Our objective is to detect exact and approximate duplicates within identity records, in order to attain a better quality of information and to permit cross-linkage among stand-alone and clustered databases. Furthermore, we need to assist human decision making, by computing a value reflecting identity proximity.MethodsThe proposed method is in three steps. The first step is to standardise and to index elementary identity fields, using blocking variables, in order to speed up information analysis. The second is to match similar pair records, relying on a global similarity value taken from the Porter-Jaro-Winkler algorithm. And the third is to create clusters of coherent related records, using graph drawing, agglomerative clustering methods and partitioning methods.ResultsThe batch analysis of 300,000 "supposedly" distinct identities isolates 240,000 true unique records, 24,000 duplicates (clusters composed of 2 records) and 3,000 clusters whose size is greater than or equal to 3 records.ConclusionDuplicate-free databases, used in conjunction with relevant indexes and similarity values, allow immediate (i.e.: real-time) proximity detection when inserting a new identity.

Thyroid Disease and Increased Cardiovascular Risk

The effects of thyroid dysfunction are thought to be reversible on restoration of euthyroidism, but postmortem and epidemiologic data suggest that subclinical or treated thyroid disease is associated with increased vascular risk. In order to determine the extent of this risk, and to explore whether the nature and/or treatment of thyroid disease are critical in this relationship, we used medical record linkage to match patients with treated thyroid disease of various etiologies with routinely collected national inpatient and daycase hospital discharge records and death records, and assessed the number of hospitalizations from cardiovascular or cerebrovascular disease or death in patients with thyroid disease and control patients. Patients treated for Graves' disease had more hospitalizations from cardiovascular disease than controls (relative risk, 1.42; 95% confidence interval, 1.20 to 1.67; p < 0.001). Toxic multinodular goiter was also associated with significantly higher rates of cardiovascular disease (relative risk, 1.50; 95% confidence interval, 1.11 to 2.02; p = 0.008). Patients with Hashimoto's thyroiditis aged over 50 years had a threefold increase in cardiovascular admissions compared to controls (23.5% and 6.5%, respectively; 95% confidence interval for difference, 6.0% to 27.9%; p = 0.003). Thus, different forms of thyroid disease were associated with increased long-term vascular risk despite restoration of euthyroidism. The mechanisms that mediate this risk are unclear but may not involve thyroid hormone abnormality.

The SCAR‐3 study: 5‐year adhesion‐related readmission risk following lower abdominal surgical procedures

The Surgical and Clinical Adhesions Research (SCAR) and SCAR-2 studies demonstrated that the burden of adhesions following lower abdominal surgery is considerable and appears to remain unchanged despite advances in strategies to prevent adhesions. In this study, we assessed the adhesion-related readmission risk directly associated with common lower abdominal surgical procedures, taking into account the effect of previous surgery, demography and concomitant disease. Data from the Scottish National Health Service medical record linkage database were used to assess the risk of an adhesion-related readmission following open lower abdominal surgery during April 1996-March 1997. Patients undergoing lower abdominal surgery (excluding appendicectomy) had a 5% risk of readmission directly related to adhesions in the 5 years following surgery. Appendicectomy was associated with a lower rate of readmission (0.9%), but contributed over 7% of the total lower abdominal surgery patient readmission burden. Panproctocolectomy (15.4%), total colectomy (8.8%) and ileostomy surgery (10.6%) were associated with the highest risk of an adhesion-related readmission. Overall, the risk of readmission was doubled in patients who had undergone abdominal or pelvic surgery within 5 years of the incident operation. A higher risk of readmission was also recorded in patients aged < 60 years compared with those aged > or = 60 yrs. The effect of gender was assessed. However, as the surgical codes used were found to be skewed towards women, these data have not been reported. Readmission risk was slightly higher in patients with concomitant peritonitis compared with patients without peritonitis. In contrast, Crohn's disease had no effect on risk. Patients with colorectal cancer had a lower risk of adhesion formation. However, this may have been due to the type of surgery performed in this patient group. The identification of high-risk patient subgroups may assist in effectively targeting adhesion-prevention strategies and the proffering of preoperative advice on adhesion risk.

The Avon Longitudinal Study of Parents and Children (ALSPAC)--study design and collaborative opportunities

The Avon Longitudinal Study of Parents and Children (ALSPAC) is a population-based study designed to understand the ways in which the physical and social environment interact, over time, with the genotype to affect health, behaviour and development. This information about causal interactions in common diseases and disorders will provide the basis on which future preventive interventions can be tested, especially in regard to people with specific genotypes. Whilst ALSPAC builds on data and hypotheses generated by earlier general population cohorts, its design offers special advantages, the most important being: enrolment in early pregnancy; banking of DNA from the children and parents (permitting genetic transmission tests and transgenerational imprinting studies); diverse physical, psychological and environmental measures; one geographical base (permitting medical record and school linkage; environmental measures in the home, and clinics for direct examination); annual hands-on examinations of the full cohort since age 7. The 14 541 enrolled pregnancies (expected date of delivery 1 April 1991 to 31 December 1992) represented about 85% of the eligible population. The 13 971 children who were still alive at age 12 months have been followed principally by questionnaires completed by a parent. Twelve years on, questionnaires are still being sent out to 11 300 families. The average questionnaire response rate from mothers is 79%. From age 7, annual examinations of both 'physical' and 'psychological' aspects have also been conducted on the children (approximately 8000 attend each 'clinic'). The quality of the data has proved high in validation studies and is maintained by collaboration with experts in each specialist field. The high participation rate ensures a viable study well into the future.

Colorectal surgery: the risk and burden of adhesion‐related complications

Adhesions are associated with serious medical complications. This study examines the real-time burden of adhesion-related readmissions following colorectal surgery and assesses the impact of previous surgery on adhesion-related outcomes. The study used data from the Scottish National Health Service Medical Record Linkage Database to identify three cohorts of patients who had undergone open colorectal surgery during the financial years 1996-97, 1997-98 and 1998-99. Each cohort was followed up for at least 2 years and the number and category of adhesion-related readmissions was recorded. The influence of any previous operations on adhesion-related readmissions was also determined by performing a subanalysis within the 1996-97 cohort of patients who had no record of abdominal surgery within either the previous 5 or 15 years. The relative risk of adhesion-related readmissions was also assessed. In the 1996-97 cohort, 9.0% of patients were readmitted within a year after surgery; 2.1% had complications directly related to adhesions and 6.9% had complications that were possibly related. After 4 years, 19.0% of patients were readmitted for reasons directly or possibly related to adhesions. Many patients were readmitted on more than one occasion and the relative risk of adhesion-related complications was 29.7 per 100 initial procedures over 4 years. In the subgroups that had no record of abdominal surgery within the previous 5 or 15 years, the relative risks of adhesion-related complications were 24.8% and 23.5%, respectively. There was no change in the rate of adhesion-related readmissions following colorectal surgery between 1996 and 1999. Colorectal surgery is associated with a considerable rate of adhesion-related readmissions. Preventative measures should be considered to reduce this risk.

Carotid endarterectomy in Scotland: 1981-1996.

Our aim was to assess the geographical, social and hospital variation in carotid endarterectomy (CEA) over time in Scotland using routinely generated data, and to assess the outcome of CEA in terms of the 7 day case-fatality. A descriptive, retrospective study was carried out using computerized medical records at national level of all patients in Scotland who had a CEA during 1981-1996. Medical record linkage was used to identify the study population with the index event specified as CEA. The annual rate of CEA in Scotland and length of hospital stay for the procedure were determined. The geographical distribution of CEA, and the social circumstances of these CEA patients were described. Hospitals where CEAs were performed were evaluated in terms of high-, medium- and low-volume hospitals. A total of 2,892 CEA patients were identified for the period 1981-1996. The CEA rate increased from 1.2/100,000 (1989) to 8.6/100,000 (1996) with a maximum of 19/100,000 in Tayside Health Board (1994). Most of the CEAs in Scotland after the publication of the trials were in a small number of 'high'-volume hospitals (> 50 operations/year). There was no gender inequality in the provision of CEA in Scotland but substantial social and regional variation was observed. The 7 day operative mortality of 1.5 per cent was higher than that observed in the randomized trials. CEA in Scotland is performed now at about the expected rate, but there are still probably too many operations being carried out in low-volume hospitals. Length of stay has decreased over time. The operative mortality, however, was higher than in the randomized trials.

Oral anticoagulation and risk of death: a medical record linkage study

Abstract Objective: To study how mortality varies with different degrees of anticoagulation reflected by the international normalised ratio (INR). Design: Record linkage analysis with death hazard estimated as a continuous function of INR. Data sources: 46 anticoagulation clinics in Sweden with computerised medical records. Subjects: Records for 42 451 patients, 3533 deaths, and 1.25 million INR measurements. Main outcome measures: Mortality from all causes and from intracranial haemorrhage. Results: Mortality from all causes of death was strongly related to level of INR. Minimum risk of death was attained at 2.2 INR for all patients and 2.3 INR for patients with mechanical heart valve prostheses. A high INR was associated with an excess mortality: with an increase of 1 unit of INR above 2.5, the risks of death from cerebral bleeding (149 deaths) and from any cause were about doubled. Among patients with an INR of 3.0, 1069 deaths occurred within 7 weeks; if the risk coincided with that with an INR of 2.9, the expected number of deaths would have been 569. Thus at least 500 deaths were associated with a high INR value, but not necessarily caused by the treatment. Conclusions: The excess mortality associated with high INR values supports the use of less intensive treatment and a small therapeutic window, with INR close to 2.2-2.3 irrespective of the indication for anticoagulant treatment. More preventive actions should be taken to avoid episodes of high INR. What is already known on this topic The optimal degree of anticoagulation (expressed as the international normalised ratio (INR)) for different indications is still unclear, but the increased risk of death due to bleeding at high INR values is well known What this study adds This large study of medical records from anticoagulation clinics in Sweden confirmed the substantial excess mortality at high INR values and indicated optimal treatment to be in a small therapeutic window with INR close to 2.2-2.3, irrespective of the indication for anticoagulant treatment

Oral anticoagulation and risk of death: a medical record linkage study

Objective: To study how mortality varies with different degrees of anticoagulation reflected by the international normalised ratio (INR).Design: Record linkage analysis with death hazard estimated as a continuous function of...

Coronary atherosclerosis in diabetes mellitus: A population-based autopsy study

ObjectivesThe study was conducted to test the hypothesis that the prevalence of coronary atherosclerosis is greater among diabetic than among nondiabetic individuals and is similar for diabetic individuals without clinical coronary artery disease (CAD) and nondiabetics with clinical CAD. BackgroundPersons with diabetes but without clinical CAD encounter cardiovascular mortality similar to nondiabetic individuals with clinical CAD. This excess mortality is not fully explained. We examined the association between diabetes and coronary atherosclerosis in a geographically defined autopsied population, while capitalizing on the autopsy rate and medical record linkage system available via the Rochester Epidemiology Project, which allows rigorous ascertainment of coronary atherosclerosis, clinical CAD, and diabetes. MethodsUsing two measures, namely a global coronary score and high-grade stenoses, the prevalence of atherosclerosis was analyzed in a cohort of autopsied residents of Rochester, Minnesota, age 30 years or older at death, while stratifying on diabetes, clinical CAD diagnosis, age, and gender. ResultsIn this cohort, diabetes was associated with a higher prevalence of atherosclerosis. Among diabetic decedents without clinical CAD, almost three-fourths had high-grade coronary atherosclerosis and more than half had multivessel disease. Without diabetes, women had less atherosclerosis than men, but this female advantage was lost with diabetes. Among those without clinical CAD, diabetes was associated with a global coronary disease burden and a prevalence of high-grade atherosclerosis similar to that observed among nondiabetic subjects with clinical CAD. ConclusionsThese findings provide mechanistic insights into the excess risk of clinical CAD among diabetic individuals, thereby supporting the need for aggressive prevention of atherosclerosis in all diabetic individuals, irrespective of clinical CAD symptoms.

Coronary disease surveillance in Olmsted County objectives and methodology

The community surveillance study of coronary heart disease (CHD) in Olmsted County, MN, is designed to estimate trends in myocardial infarction (MI) incidence, case fatality rate, and CHD mortality, while including all ages. A distinctive feature of this study is its ability to capture longitudinal data before and after index events via the medical record linkage system of the Rochester Epidemiology Project. The goal of this report is to describe the methods implemented to measure CHD trends, the implications of including elderly individuals on MI ascertainment and trends in prior CHD among persons with incident MI. The methods are based on standardized criteria involving the review of death certificate information and hospital records to identify CHD deaths, and incident MIs in Olmsted County. The medical record linkage system in place under the auspices of the Rochester Epidemiology Project was used to ascertain antecedent CHD and outcomes. Hospitalized MIs were screened from sampled events coded ICD9 codes 410–414 and classified using enzyme values, cardiac pain, and ECG coding. After screening 5,042 records, a cohort of 1,658 validated incident MIs was assembled 35% (575) among persons aged 75 years or greater. The proportion of MIs validated with cardiac pain and enzymes without Minnesota ECG coding was lower among the elderly than among persons less than 75 years of age (35 vs. 29%, respectively; P < .001). The proportion of events validated without requiring ECG coding decreased over time in both age strata (P for trend .001). Reliability analyses indicated excellent agreement in event classification. More than half of the incident MIs did not have antecedent CHD, and this proportion increased overtime. These data indicate that the elderly contribute approximately one-third of the cases of incident MI, underscoring the importance of including all ages to fully characterize the burden of CHD. Cases among elderly persons more frequently require ECG coding for validation, but standardized ascertainment procedures are feasible and reliable in all age groups. More than half of the incident MIs occurred among persons with no prior CHD, and this proportion increased over time. The combination of standardized methodology and of the longitudinal data via the record linkage system of the Rochester Epidemiology Project will allow reliable measures of CHD trends and help define preventive strategies.

Postoperative adhesions: ten-year follow-up of 12,584 patients undergoing lower abdominal surgery.

Postoperative adhesions are a significant problem after colorectal surgery. However, the basic epidemiology and clinical burden are unknown. The Surgical and Clinical Adhesions Research Study has investigated the scale of the problem in a population of 5 million. Validated data from the Scottish National Health Service Medical Record Linkage Database were used to define a cohort of 12,584 patients undergoing open lower abdominal surgery in 1986. Readmissions for potential adhesion-related disease in the subsequent ten years were analyzed. The methodology was conservative in interpreting adhesion-related disease. In the study cohort 32.6 percent of patients were readmitted a mean of 2.2 times in the subsequent ten years for a potential adhesion-related problem. Although 25.4 percent of readmissions were in the first postoperative year, they continued steadily throughout the study period. After open lower abdominal surgery 7.3 percent (643) of readmissions (8,861) were directly related to adhesions. This varied according to operation site: colon (7.1 percent), rectum (8.8 percent), and small intestine (7.6 percent). The readmission rate was assessed to provide an indicator of relative risk of adhesion-related problems after initial surgery. The overall average rate of readmissions was 70.4 per 100 initial operations, with 5.1 directly related to adhesions. This rose to 116.4 and 116.5, respectively, after colonic or rectal surgery-with 8.2 and 10.3 directly related to adhesions. There is a high relative risk of adhesion-related problems after open lower abdominal surgery and a correspondingly high workload associated with these readmissions. This is influenced by the initial site of surgery, colon and rectum having both the greatest impact on workload and highest relative risk of directly adhesion-related problems. The study provides sound justification for improved adhesion prevention strategies.

The Burden of Adhesions

Just how much trouble do postoperative adhesions cause? Using data from the Scottish National Health Service Medical Record Linkage Database, these researchers analyzed 12,584 patients who had undergone open lower abdominal surgery in 1986. A conservative methodologic …

The impact of adhesions on hospital readmissions over ten years after 8849 open gynaecological operations: an assessment from the Surgical and Clinical Adhesions Research Study.

To investigate the epidemiology of, and the clinical burden related to, adhesions following gynaecological surgery. The Scottish National Health Service Medical Record Linkage Database was used to define a cohort of 8849 women undergoing open gynaecological surgery in 1986. All readmissions for potential adhesion related disease in the subsequent 10 years were reviewed. Readmissions and the degree of adhesion involvement gave an indication of clinical burden and workload. The rate of readmission following the initial surgery determined the relative risk of disease related to adhesions. Two hundred and forty-five (4.5%) of 5433 readmissions following open gynaecological surgery were directly related to adhesions. 34.5% of patients were readmitted, on average 1.9 times, for a problem potentially related to adhesions or for further intra-abdominal surgery that could be complicated by adhesions. Readmissions related to adhesions continued throughout the 10 year period of the study. The overall rate of readmission was 64.0/100 initial operations. For readmissions directly related to adhesions, the rate was 2.9/100 initial operations. Operations on the ovary had the highest rate directly related to adhesions (7.5/100 initial operations), with an overall rate of readmission of 106.4/100 initial operations. Despite the conservative approach taken in this study, the clinical burden, workload and relative risk of readmissions related to adhesions following open gynaecological surgery was considerable. Post-operative adhesions have important consequences for patients, surgeons and the healthcare system. These results emphasise the need for more effective strategies to prevent adhesions.

Risk factors for nursing home placement in a population-based dementia cohort.

To examine risk factors for nursing home placement in a population-based dementia cohort. The Mayo Clinic Medical Records linkage system was used to identify all patients with onset of dementia between 1980 and 1984. The patient group included 314 cases who met DSM-III-R criteria for dementia, including 220 cases who were community dwelling at onset. All dementia patients were followed until death. A control group included 323 patients who did not, at any point, meet DSM-III-R criterion for dementia. The groups were initially matched on age, gender, and year of initial registration. Demographic, medical, social, and functional predictors were examined as static and time-dependent risk factors for nursing home placement in the initial community-dwelling subgroups, using stepwise Cox regression modeling. Of the 314 dementia patients, 282 took residence in licensed skilled nursing homes for at least 6 weeks, suggestive of custodial care, at some point during the course of their illness. In the control group, 162 of the 323 people required nursing home placement. Within controls, the predictor variables of time to nursing home placement included initial age, being divorced, living in a townhome, apartment or assisted living apartment, change in Charlson comorbidity score, and change in amount of daily assistance required. Within the dementia sample, seven predictors were eventually determined to be associated with time to nursing home placement. These included total number of years of education, age at onset of dementia, being single, living in a retirement or supervised apartment at onset, change in Charlson comorbidity score, and a change in the amount of daily assistance required. Cumulative incidence of placement was 90% in the dementia cohort and 50% in the controls. Certain variables seem to impact time to nursing home placement in all older persons, whether they have dementia or not. Among these are age, living in assisted living settings, increasing comorbidity scores, and increasing need for functional assistance. Certain additional factors may have a specific impact in dementia. Among these is education, which seems to provide a protective effect. These predictors may be important covariates in clinical dementia studies that include time to nursing home placement as an outcome variable.

Is Surgery with Anesthesia an Independent Risk Factor for Stroke? Wong et al. (page 425)

Although certain surgeries, such as cardiac, neurologic, and vascular procedures, and risk factors such as cigarette smoking, male gender, and history of transient ischemic attacks have been established as associated with increased incidence of perioperative stroke, no one has evaluated surgery and anesthesia alone as a possible independent risk factor for development of stroke.Using a medical record linkage system to the Rochester Epidemiology Project, Wong et al. retrieved records of 1,455 people who had an incident (first-time) of ischemic stroke from 1960 to 1984. An equal number of age-and gender-matched controls were identified from the same database. With the additional usage of the Mayo Surgical Information Retrieval System, the team then identified all stroke cases in which surgery with either general anesthesia or central neuroaxis blockade had been performed within 1 yr before the stroke date.A conditional logistic regression model was used to estimate the odds of stroke after surgery and general anesthesia. The team identified 59 patients and 17 controls who underwent surgery within 30 days before their stroke or index date. After adjusting for previously identified risk factors (male gender, cigarette smoking, history of transient ischemic attacks, among others) the team found that undergoing surgery 30 days before the index date was an independent risk factor for stroke. The risk of perioperative ischemic stroke was increased even after general, non–high-risk surgeries. The mechanisms underlying this increased risk, of course, require further investigation.