Variety of endosomal TLRs and Interferons (IFN-α, IFN-β, IFN-γ) expression profiles in patients with SLE, SSc and MCTD

We investigated Toll-like receptor (TLR)-3/-7/-8/-9 and interferon (IFN)-α/β/γ mRNA expression in whole blood and serum IFN-α/β/γ levels in patients with mixed connective tissue disease (MCTD), systemic lupus erythematosus (SLE) and systemic sclerosis (SSc) and in healthy subjects to assess the association between the TLR-IFN expression and severity of and susceptibility to diseases, and identify potential biomarkers. Expression of the IFN-γ, TLR-3 and TLR-8 was detected only in SLE patients. TLR-7, IFN-α and IFN-β expression was highest in SLE, while TLR-9 expression was highest in SSc patients. In SLE and MCTD patients a strong correlation was observed between TLR-7 and IFN-α expression and IFN-β and IFN-α expression. In MCTD patients, negative correlation between IFN-α and TLR-9 and TLR-7 and TLR-9 was revealed. TLR-9 expression in anti-U1-70k-negative, anti-C negative and anti-SmB-negative MCTD patients was higher than in MCTD-positive patients. We observed negative correlations between serum IFN-α levels and TLR-7 expression and C3 and C4 levels in SLE patients. In SLE patients we observed that with increased IFN-γ, TLR-3 and TLR-8 expression increased the value of C3 and C4. Our results confirmed that the endosomal TLR-IFN pathway seems to be more important in SLE than in MCTD or SSc, and that IFN-α and IFN-β may be possible biomarkers for SLE.

THU0277 THE EXPRESSION OF IFNΑ, INFΒ AND INFΓ AND SERUM LEVELS OF THOSE CYTOKINES IN SJÖGREN’S SYNDROME PATIENTS

Background:

In the pathogensis of autoimmune mediated diseases, such as Sjögren’s syndrome (SS), interferons (IFN) and IFN pathway activation play a vital role.

Objectives:

We planned to assess IFNα, INFβ and INFγ expression and IFNs serum levels in SS patients and correlation of these parameters with: autoantibodies specific for SS, serum concentration of C3, C4 component of complement (C3, C4), rheumatoid factor (RF), gammaglobulins, focus score (FS) and eye dryness symptoms.

Methods:

Whole blood RNA was isolated from 77 SS patients [F91%vsM9%]; mean age 49,69±15.36; SS diagnosis according to EULAR/ACR 2016 criteria. The analysis of INFα, - β and - γ expression levels was based on validated TaqMan probes by ΔCT methods. Serum concentrations of rheumatoid factor (RF), C3- and C4 complement components (mg/dL) and gammaglobulins (g/dL), were assessed. Anti-Ro/SSA and/or anti-La/SSB autoantibodies were assessed by semiquantitative immunoblotting evaluation. The eye dryness and keratoconjunctivitis sicca were confirmed with Schirmer’s test (score of less than 5 mm/5’) and the ocular staining score (OSS) using lissamine green and fluorescein staining. The biopsy of minor salivary gland was performed with the histopathological evaluation of FS. The study was approved by the Bioethics Committee. Differences between groups of patients were determined using non-parametric Mann-Whitney U test or Kruskall-Wallis test with Dunn’s post hoc. Correlations were determined using non-parametric Spearman test. The level of statistical significance was set at p < 0.05.

Results:

IFNβ had the highest expression levels among IFNs and IFNβ serum concentrations were higher than those of IFNα and -γ. In cases with high IFNβ serum concentration lower IFNβ expression was observed. There was a highly significant correlation between IFNα and IFNβ expression (r =0.6;p=0.001). IFNβ expression (p=0.059) was higher in the group of younger (<45 y.o.) patients (n= 23; 29.9%) as compared to the group of older individuals (at least 45 y.o.). In patients with SS-A / Ro antibodies with strong antigen binding affinity (3) IFN β expression and IFNβ serum levels were highest of all IFNs. The presence of anti La/SS-B antibodies was associated with the increased IFNβ expression while not with the increased IFNβ serum concentration. In terms of IFNα expression and protein level, RF(+)patients had average higher values compared to RF(-) patients. The average mRNA level of IFNα was about 3 times lower in patients with low C3 serum concentration compared to patients with normal C3 serum concentration values. IFNβ mRNA level was 2.5 times lower in patients with low Schirmer’s test (<5mm/5’) in comparison to patients with Schirmer’s test>5mm/5’; Schirmer’s test <5mm/5’ was associated with higher IFNβ serum concentration.

Conclusion:

Type I IFN signature predominates in the peripheral blood of studied patients. Presented results confirmed the pivotal role of type I IFN in the disease process. The serum concentration of IFNβ and the expression of IFNβ were the highest values of those parameters for cytokines assessed in this study. A positive correlation between IFNα and IFNβ mRNA levels has been observed.

Disclosure of Interests:

None declared

AB0603 PDGFΑ AS A POTENTIAL BLOOD MARKER IN DSSC

Background:

PDGF is a potential important factor in the pathogenesis of scleroderma. PDGF is almost undetectable in healthy skin or lung. Immunohistochemical studies have revealed increased presence of PDGFαand PDGFβreceptors in scleroderma skin biopsies.

Objectives:

The aim of this study was to determine the mRNA level ofIFNα1,IL-4, TGFβ1,TGFβ2,PDGFα,PDGFβ,TNFαin whole blood in SSc patients in the aspect of clinical

Methods:

A group of 14 patients (50% were women) with systemic sclerosis based on EULAR / ACR 2013 criteria was included in the study. The modified Rodnan Skin Score (mRSS) was evaluated by same assistant at the beginning of the study and six months later. DLCO, HRCT, echocardiography and NFC were measured.

Gene expression was determined using validated TaqMan probes in qPCR. Constitutive mRNA level of selected genes was analyzed using ΔCt method. Comparison between different groups of patients was determined using non-parametric Mann-Whitney U test. Correlation was analyzed using non-parametric Spearman test.

Results:

The mean age of the patients was 60 ± 15.66. 100% of patients had organ involvement as pulmonary fibrosis. 78% - had active changes -features of ground glass.

64% of patients had mild mRSS-1-10 skin involvement, 36% had moderate to severe skin involvement. In SSc patientsTGFβ1andIFNα1revealed the highest level of expression in comparison to other analyzed genes. Additionally, very high and significant correlation betweenTNFαandTGFβ1(r=0.7 p=0.004) has been noted. High and significant correlation between mRNAPDGFβandTNFαlevels have been observed. We did not reveal significant differences in analyzed genes expression when compare limited and diffuse SSc. Nevertheless, patients with dSSc were characterized by higher level ofIFNα1(almost 2 times) andTGFβ1. On the border of significance higherPDGFαmRNA level was observed in dSSc patients when compared to lSSc. AveragePDGFαexpression is higher in SSc patients with Scl70 positive than than in patients without Scl70 (p=0.04).

In the aspect of clinical parameters, patients with ESR ≤12 mm/h revealed almost 6 times higher level ofIFNα1(p=0.01) in comparison to the patient with ESR>12mm/h.

Patients with mRSS above10 points revealed significantly higher ofPDGFαexpression in comparison to patients with mRSS ≤10 (p=0.04). In these group of patients CRP and ESR were not different significantly.

In the case of patients with active fibrosis (ground glass) in HRCTIFNα1expression was almost 2.5-times higher than in patients with HRCT non-active. Significantly higherPDGFα has been revealed in patients with active HRCT when compared with patient with non-active HRCT. Nevertheless these two groups did not differ in ESR or OB parameter.

SSc patients in active phase of NFC revealed almost 3 times higher level ofIFNα1expression in comparison with the patient in late phase.

Conclusion:

The mRNA level ofPDGFαmay be a potential blood marker to predict worse prognosis in scleroderma

References:

[1]Rheumatology (Oxford). 2008 Oct;47 Suppl 5:v2-4. doi: 10.1093/rheumatology/ken265.

Role of PDGF in fibrotic diseases and systemic sclerosis.

Trojanowska M1.

Acknowledgments:

No

Disclosure of Interests:

None declared

Record Linkage Strategies: Part II. Portable Software and Deterministic Matching

Software to perform record linkage should have several characteristics: (1) portability in being able to function with researchers’ current arrangement of computer systems and languages, (2) flexibility in handling different linkage strategies, and (3) low cost in both computer time and researchers’ efforts. A linkage package (LINKS) is described which satisfies these criteria; LINKS provides tools for both deterministic and probabilistic linkage as well as test modules for assessing data quality and structure. Because each linkage project is different, the modular nature of the software allows for better control of the programming process and development of unique strategies. Since the user provides the weights and decision rules, he may modify data between steps and/or develop extra steps to supplement the basic modules. In two information-rich linkage projects involving California AIDS data, LINKS identified mortality using deterministic approaches and permitted comparisons with other software and strategies. Flexible software and a deterministic approach would have eliminated the expensive key entry used to add full names and social security numbers as additional identifiers to one of the California data files.

Record Linkage Strategies

Record linkage techniques can help identify the same patient for matching diverse files (hospital discharge abstracts, insurance claims, registries, Vital Statistics data) which contain similar identifiers. Prior knowledge of whether a linkage is feasible is important to prevent wasted effort (additional data collection or data manipulation), which decreases the cost-effectiveness of the linkage. Using examples generated by linking the Manitoba Health Services Commission data with Vital Statistics files, a method of estimating the information in each data set is presented first. Further, the feasibility of several different record linkage strategies is described and tested, given varying amounts of information. At the margin, relatively small amounts of information (having just one more variable to match with) can make a great difference. Probabilistic linkage’s great advantage was found in those situations where only a moderate amount of extra information was available.

By using the above techniques when working with one or both files in a proposed record linkage project, a much more informed judgement can now be made as to whether a linkage will or will not work. In facilitating record linkage, flexibility of both software and the strategy for matching is very important.

Cell and region specificity of Aryl hydrocarbon Receptor (AhR) system in the testis and the epididymis

Aryl hydrocarbon receptor (AhR) plays multiple important functions in adaptive responses. Exposure to AhR ligands may produce an altered metabolic activity controlled by the AhR pathways, and consequently affect drug/toxin responses, hormonal status and cellular homeostasis. This research revealed species-, cell- and region-specific pattern of the AhR system expression in the rat and human testis and epididymis, complementing the existing knowledge, especially within the epididymal segments. The study showed that AhR level in the rat and human epididymis is higher than in the testis. The downregulation of AhR expression after TCDD treatment was revealed in the spermatogenic cells at different stages and the epididymal epithelial cells, but not in the Sertoli and Leydig cells. Hence, this basic research provides information about the AhR function in the testis and epididymis, which may provide an insight into deleterious effects of drugs, hormones and environmental pollutants on male fertility.

TGF-β1 gene polymorphism in renal transplant patients with and without gingival overgrowth

BACKGROUND

The incidence of gingival overgrowth among renal transplant patients treated with cyclosporine A ranges from 13% to 84.6%, and the overgrowth is not only esthetic but also a medical problem. We studied the determination of association between TGF-β1 (TGFB1) gene polymorphism and gingival overgrowth in kidney transplant patients medicated with cyclosporin A.

METHODS

Eighty-four kidney transplant patients with gingival overgrowth and 140 control transplant patients without overgrowth were enrolled into the case control study. TGFB1 polymorphism was determined using the PCR-RFLP assay for +869T > C in codon 10 and +915G > C in codon 25 as well as TaqMan real-time PCR assays for promoter -800G>A and -509C > T SNPs.

RESULTS

In kidney transplant patients suffering from gingival overgrowth, mean score of gingival overgrowth was 1.38 ± 0.60, whereas in control subjects it was 0.0. The patients with gingival overgrowth were characterized by similar distribution of TGFB1 genotypes and allele in comparison to subjects without gingival overgrowth. Among 16 potentially possible haplotypes of TGFB1 gene, only four were observed in the studied sample of kidney transplant patients: G_C_T_G, G_T_C_G, G_C_C_C, and A_C_T_G, with similar frequency in patients with and without gingival overgrowth.

CONCLUSION

No association between the TGFB1 gene polymorphism and gingival overgrowth was revealed in kidney transplant patients administered cyclosporine A.

Evaluation of mutagenic and antimutagenic properties of some bioactive xanthone derivatives using Vibrio harveyi test

AIMS

Drug safety evaluation plays an important role in the early phase of drug development, especially in the preclinical identification of compounds' biological activity. The Vibrio harveyi assay was used to assess mutagenic and antimutagenic activity of some aminoalkanolic derivatives of xanthone (1-5), which were synthesized and evaluated for their anticonvulsant and hemodynamic activities.

METHODS AND RESULTS

A novel V. harveyi assay was used to assess mutagenic and antimutagenic activity of derivatives of xanthone 1-5. Two V. harveyi strains were used: BB7 (natural isolate) and BB7M (BB7 derivative containing mucA and mucB genes on a plasmid pAB91273, products of these genes enhance error-prone DNA repair). According to the results obtained, the most beneficial mutagenic and antimutagenic profiles were observed for compounds 2 and 3. A modification of the chemical structure of compound 2 by the replacement of the hydroxy group by a chloride improved considerably the antimutagenic activity of the compound. Thus, antimutagenic potency reached a maximum with the presence of tertiary amine and chloride atom in the side chain.

CONCLUSIONS

Among the newly synthesized aminoalkanolic derivatives of xanthone with potential anticonvulsant properties, there are some compounds exhibiting in vitro antimutagenic activity. In addition, it appears that the V. harveyi assay can be applied for primary mutagenicity and antimutagenicity assessment of compounds.

SIGNIFICANCE AND IMPACT OF THE STUDY

The obtained preliminary mutagenicity and antimutagenicity results encourage further search in the group of amino derivatives of xanthone as the potential antiepileptic drugs also presenting some antimutagenic potential. Furthermore, V. harveyi test may be a useful tool for compounds safety evaluation.

Small-area variations and sociodemographic correlates for the incidence of Crohn's disease and ulcerative colitis

The objectives of this study were to describe variations in the incidence of inflammatory bowel disease (IBD) within the Canadian province of Manitoba and to analyze sociodemographic factors associated with these variations. The authors used the Manitoba Health insurance databases to measure incidence rates of Crohn's disease and ulcerative colitis for each of 52 postal areas in Manitoba, in 1987-1996. The sociodemographic characteristics of the postal areas were based on data from the 1996 Canadian census. The overall incidence rates of Crohn's disease and ulcerative colitis were identical-15.6 per 100,000. Both diseases showed substantial geographic variation, with incidence rates differing significantly from the provincial average in 15 postal areas for Crohn's disease and in 13 postal areas for ulcerative colitis. There was a significant geographic correlation in the incidence of Crohn's disease and ulcerative colitis (r = 0.49, p < 0.001). The incidence of IBD was higher in urban areas (incidence rate ratio (IRR) = 1.21, 95% confidence interval (CI): 1.00, 1.45). Aboriginal Canadians had significantly lower rates of both Crohn's disease (IRR = 0.11, 95% CI: 0.05, 0.22) and ulcerative colitis (IRR = 0.57, 95% CI: 0.42, 0.79). A higher incidence of IBD was ecologically associated with a higher average family income, a lower proportion of immigrant and Aboriginal Canadian populations, and a smaller average family size.

Cancer risk in patients with inflammatory bowel disease: a population-based study

BACKGROUND

The objective of the current study was to determine the incidence of cancer among persons with inflammatory bowel disease (IBD) and to compare these incidence rates with those of the non-IBD population using population-based data from the administrative claims data of Manitoba's universal provincial insurance plan (Manitoba Health).

METHODS

IBD patients were matched 1:10 to randomly selected members of the population without IBD based on year, age, gender, and postal area of residence. The incidence of cancer was determined by linking records from the IBD and non-IBD cohorts with the comprehensive Cancer Care Manitoba registry. Incidence rates and rate ratios (IRR) were calculated based on person-years of follow-up (Crohn's disease = 21,340 person-years and ulcerative colitis [UC] = 19,665 person-years) for 1984-1997.

RESULTS

There was an increased IRR of colon carcinoma for both Crohn disease patients (2.64; 95% confidence interval [95% CI], 1.69-4.12) and UC patients (2.75; 95% CI, 1.91-3.97). There was an increased IRR of rectal carcinoma only among patients with UC (1.90; 95% CI, 1.05-3.43) and an increased IRR of carcinoma of the small intestine only in Crohn disease patients (17.4; 95% CI, 4.16-72.9). An increased IRR of extraintestinal tumors was observed only for the liver and biliary tract in both Crohn disease patients (5.22; 95% CI, 0.96-28.5) and UC patients (3.96; 95% CI, 1.05-14.9). There was an increased IRR of lymphoma for males with Crohn disease only (3.63; 95% CI, 1.53-8.62), and this finding did not appear to be related to use of immunomodulatory therapy. Compared with controls, Crohn's disease was associated with an increased risk of cancer overall, but UC was not.

CONCLUSIONS

There appear to be similar increased risks for developing colon carcinoma and hepatobiliary carcinoma among patients with Crohn disease and UC. There is an increased risk of developing rectal carcinoma in UC patients, an increased risk of developing carcinoma of the small bowel in Crohn disease patients, and an increased risk of developing lymphoma among males with Crohn disease.

Cancer risk in patients with inflammatory bowel disease: a population-based study

BACKGROUND

The objective of the current study was to determine the incidence of cancer among persons with inflammatory bowel disease (IBD) and to compare these incidence rates with those of the non-IBD population using population-based data from the administrative claims data of Manitoba's universal provincial insurance plan (Manitoba Health).

METHODS

IBD patients were matched 1:10 to randomly selected members of the population without IBD based on year, age, gender, and postal area of residence. The incidence of cancer was determined by linking records from the IBD and non-IBD cohorts with the comprehensive Cancer Care Manitoba registry. Incidence rates and rate ratios (IRR) were calculated based on person-years of follow-up (Crohn's disease = 21,340 person-years and ulcerative colitis [UC] = 19,665 person-years) for 1984-1997.

RESULTS

There was an increased IRR of colon carcinoma for both Crohn disease patients (2.64; 95% confidence interval [95% CI], 1.69-4.12) and UC patients (2.75; 95% CI, 1.91-3.97). There was an increased IRR of rectal carcinoma only among patients with UC (1.90; 95% CI, 1.05-3.43) and an increased IRR of carcinoma of the small intestine only in Crohn disease patients (17.4; 95% CI, 4.16-72.9). An increased IRR of extraintestinal tumors was observed only for the liver and biliary tract in both Crohn disease patients (5.22; 95% CI, 0.96-28.5) and UC patients (3.96; 95% CI, 1.05-14.9). There was an increased IRR of lymphoma for males with Crohn disease only (3.63; 95% CI, 1.53-8.62), and this finding did not appear to be related to use of immunomodulatory therapy. Compared with controls, Crohn's disease was associated with an increased risk of cancer overall, but UC was not.

CONCLUSIONS

There appear to be similar increased risks for developing colon carcinoma and hepatobiliary carcinoma among patients with Crohn disease and UC. There is an increased risk of developing rectal carcinoma in UC patients, an increased risk of developing carcinoma of the small bowel in Crohn disease patients, and an increased risk of developing lymphoma among males with Crohn disease.

Cancer risk in patients with inflammatory bowel disease: a population-based study

BACKGROUND

The objective of the current study was to determine the incidence of cancer among persons with inflammatory bowel disease (IBD) and to compare these incidence rates with those of the non-IBD population using population-based data from the administrative claims data of Manitoba's universal provincial insurance plan (Manitoba Health).

METHODS

IBD patients were matched 1:10 to randomly selected members of the population without IBD based on year, age, gender, and postal area of residence. The incidence of cancer was determined by linking records from the IBD and non-IBD cohorts with the comprehensive Cancer Care Manitoba registry. Incidence rates and rate ratios (IRR) were calculated based on person-years of follow-up (Crohn's disease = 21,340 person-years and ulcerative colitis [UC] = 19,665 person-years) for 1984-1997.

RESULTS

There was an increased IRR of colon carcinoma for both Crohn disease patients (2.64; 95% confidence interval [95% CI], 1.69-4.12) and UC patients (2.75; 95% CI, 1.91-3.97). There was an increased IRR of rectal carcinoma only among patients with UC (1.90; 95% CI, 1.05-3.43) and an increased IRR of carcinoma of the small intestine only in Crohn disease patients (17.4; 95% CI, 4.16-72.9). An increased IRR of extraintestinal tumors was observed only for the liver and biliary tract in both Crohn disease patients (5.22; 95% CI, 0.96-28.5) and UC patients (3.96; 95% CI, 1.05-14.9). There was an increased IRR of lymphoma for males with Crohn disease only (3.63; 95% CI, 1.53-8.62), and this finding did not appear to be related to use of immunomodulatory therapy. Compared with controls, Crohn's disease was associated with an increased risk of cancer overall, but UC was not.

CONCLUSIONS

There appear to be similar increased risks for developing colon carcinoma and hepatobiliary carcinoma among patients with Crohn disease and UC. There is an increased risk of developing rectal carcinoma in UC patients, an increased risk of developing carcinoma of the small bowel in Crohn disease patients, and an increased risk of developing lymphoma among males with Crohn disease.

Cancer risk in patients with inflammatory bowel disease: a population-based study

BACKGROUND

The objective of the current study was to determine the incidence of cancer among persons with inflammatory bowel disease (IBD) and to compare these incidence rates with those of the non-IBD population using population-based data from the administrative claims data of Manitoba's universal provincial insurance plan (Manitoba Health).

METHODS

IBD patients were matched 1:10 to randomly selected members of the population without IBD based on year, age, gender, and postal area of residence. The incidence of cancer was determined by linking records from the IBD and non-IBD cohorts with the comprehensive Cancer Care Manitoba registry. Incidence rates and rate ratios (IRR) were calculated based on person-years of follow-up (Crohn's disease = 21,340 person-years and ulcerative colitis [UC] = 19,665 person-years) for 1984-1997.

RESULTS

There was an increased IRR of colon carcinoma for both Crohn disease patients (2.64; 95% confidence interval [95% CI], 1.69-4.12) and UC patients (2.75; 95% CI, 1.91-3.97). There was an increased IRR of rectal carcinoma only among patients with UC (1.90; 95% CI, 1.05-3.43) and an increased IRR of carcinoma of the small intestine only in Crohn disease patients (17.4; 95% CI, 4.16-72.9). An increased IRR of extraintestinal tumors was observed only for the liver and biliary tract in both Crohn disease patients (5.22; 95% CI, 0.96-28.5) and UC patients (3.96; 95% CI, 1.05-14.9). There was an increased IRR of lymphoma for males with Crohn disease only (3.63; 95% CI, 1.53-8.62), and this finding did not appear to be related to use of immunomodulatory therapy. Compared with controls, Crohn's disease was associated with an increased risk of cancer overall, but UC was not.

CONCLUSIONS

There appear to be similar increased risks for developing colon carcinoma and hepatobiliary carcinoma among patients with Crohn disease and UC. There is an increased risk of developing rectal carcinoma in UC patients, an increased risk of developing carcinoma of the small bowel in Crohn disease patients, and an increased risk of developing lymphoma among males with Crohn disease.

Cancer risk in patients with inflammatory bowel disease: a population-based study

BACKGROUND

The objective of the current study was to determine the incidence of cancer among persons with inflammatory bowel disease (IBD) and to compare these incidence rates with those of the non-IBD population using population-based data from the administrative claims data of Manitoba's universal provincial insurance plan (Manitoba Health).

METHODS

IBD patients were matched 1:10 to randomly selected members of the population without IBD based on year, age, gender, and postal area of residence. The incidence of cancer was determined by linking records from the IBD and non-IBD cohorts with the comprehensive Cancer Care Manitoba registry. Incidence rates and rate ratios (IRR) were calculated based on person-years of follow-up (Crohn's disease = 21,340 person-years and ulcerative colitis [UC] = 19,665 person-years) for 1984-1997.

RESULTS

There was an increased IRR of colon carcinoma for both Crohn disease patients (2.64; 95% confidence interval [95% CI], 1.69-4.12) and UC patients (2.75; 95% CI, 1.91-3.97). There was an increased IRR of rectal carcinoma only among patients with UC (1.90; 95% CI, 1.05-3.43) and an increased IRR of carcinoma of the small intestine only in Crohn disease patients (17.4; 95% CI, 4.16-72.9). An increased IRR of extraintestinal tumors was observed only for the liver and biliary tract in both Crohn disease patients (5.22; 95% CI, 0.96-28.5) and UC patients (3.96; 95% CI, 1.05-14.9). There was an increased IRR of lymphoma for males with Crohn disease only (3.63; 95% CI, 1.53-8.62), and this finding did not appear to be related to use of immunomodulatory therapy. Compared with controls, Crohn's disease was associated with an increased risk of cancer overall, but UC was not.

CONCLUSIONS

There appear to be similar increased risks for developing colon carcinoma and hepatobiliary carcinoma among patients with Crohn disease and UC. There is an increased risk of developing rectal carcinoma in UC patients, an increased risk of developing carcinoma of the small bowel in Crohn disease patients, and an increased risk of developing lymphoma among males with Crohn disease.

Cancer risk in patients with inflammatory bowel disease: a population-based study

BACKGROUND

The objective of the current study was to determine the incidence of cancer among persons with inflammatory bowel disease (IBD) and to compare these incidence rates with those of the non-IBD population using population-based data from the administrative claims data of Manitoba's universal provincial insurance plan (Manitoba Health).

METHODS

IBD patients were matched 1:10 to randomly selected members of the population without IBD based on year, age, gender, and postal area of residence. The incidence of cancer was determined by linking records from the IBD and non-IBD cohorts with the comprehensive Cancer Care Manitoba registry. Incidence rates and rate ratios (IRR) were calculated based on person-years of follow-up (Crohn's disease = 21,340 person-years and ulcerative colitis [UC] = 19,665 person-years) for 1984-1997.

RESULTS

There was an increased IRR of colon carcinoma for both Crohn disease patients (2.64; 95% confidence interval [95% CI], 1.69-4.12) and UC patients (2.75; 95% CI, 1.91-3.97). There was an increased IRR of rectal carcinoma only among patients with UC (1.90; 95% CI, 1.05-3.43) and an increased IRR of carcinoma of the small intestine only in Crohn disease patients (17.4; 95% CI, 4.16-72.9). An increased IRR of extraintestinal tumors was observed only for the liver and biliary tract in both Crohn disease patients (5.22; 95% CI, 0.96-28.5) and UC patients (3.96; 95% CI, 1.05-14.9). There was an increased IRR of lymphoma for males with Crohn disease only (3.63; 95% CI, 1.53-8.62), and this finding did not appear to be related to use of immunomodulatory therapy. Compared with controls, Crohn's disease was associated with an increased risk of cancer overall, but UC was not.

CONCLUSIONS

There appear to be similar increased risks for developing colon carcinoma and hepatobiliary carcinoma among patients with Crohn disease and UC. There is an increased risk of developing rectal carcinoma in UC patients, an increased risk of developing carcinoma of the small bowel in Crohn disease patients, and an increased risk of developing lymphoma among males with Crohn disease.

The incidence of deep venous thrombosis and pulmonary embolism among patients with inflammatory bowel disease: a population-based cohort study

BACKGROUND

There is an impression mostly from specialty clinics that patients with inflammatory bowel disease (IBD) have an increased risk of venous thromboembolic disorders. Our aim was to determine the incidence of deep venous thrombosis (DVT) and pulmonary embolism (PE) from a population-based database of IBD patients and, to compare the incidence rates to that of an age, gender and geographically matched population control group.

METHODS

IBD patients identified from the administrative claims data of the universal provincial insurance plan of Manitoba were matched 1:10 to randomly selected members of the general population without IBD by year, age, gender, and postal area of residence using Manitoba Health's population registry. The incidence of hospitalization for DVT and PE was calculated from hospital discharge abstracts using ICD-9-CM codes 451.1, 453.x for DVT and 415.1x for PE. Rates were calculated based on person-years of follow-up for 1984-1997. Comparisons to the population cohort yielded age-adjusted incidence rate ratios (IRR). Rates were calculated based on person-years of follow-up (Crohn's disease = 21,340, ulcerative colitis = 19,665) for 1984-1997.

RESULTS

In Crohn's disease the incidence rate of DVT was 31.4/10,000 person-years and of PE was 10.3/10,000 person-years. In ulcerative colitis the incidence rates were 30.0/10,000 person-years for DVT and 19.8/10,000 person-years for PE. The IRR was 4.7 (95% CI, 3.5-6.3) for DVT and 2.9 (1.8-4.7) for PE in Crohn's disease and 2.8 (2.1-3.7) for DVT and 3.6 (2.5-5.2) for PE, in ulcerative colitis. There were no gender differences for IRR. The highest rates of DVT and PE were seen among patients over 60 years old; however the highest IRR for these events were among patients less than 40 years.

CONCLUSION

IBD patients have a threefold increased risk of developing DVT or PE.

Cancer risk in patients with inflammatory bowel disease: a population-based study

BACKGROUND

The objective of the current study was to determine the incidence of cancer among persons with inflammatory bowel disease (IBD) and to compare these incidence rates with those of the non-IBD population using population-based data from the administrative claims data of Manitoba's universal provincial insurance plan (Manitoba Health).

METHODS

IBD patients were matched 1:10 to randomly selected members of the population without IBD based on year, age, gender, and postal area of residence. The incidence of cancer was determined by linking records from the IBD and non-IBD cohorts with the comprehensive Cancer Care Manitoba registry. Incidence rates and rate ratios (IRR) were calculated based on person-years of follow-up (Crohn's disease = 21,340 person-years and ulcerative colitis [UC] = 19,665 person-years) for 1984-1997.

RESULTS

There was an increased IRR of colon carcinoma for both Crohn disease patients (2.64; 95% confidence interval [95% CI], 1.69-4.12) and UC patients (2.75; 95% CI, 1.91-3.97). There was an increased IRR of rectal carcinoma only among patients with UC (1.90; 95% CI, 1.05-3.43) and an increased IRR of carcinoma of the small intestine only in Crohn disease patients (17.4; 95% CI, 4.16-72.9). An increased IRR of extraintestinal tumors was observed only for the liver and biliary tract in both Crohn disease patients (5.22; 95% CI, 0.96-28.5) and UC patients (3.96; 95% CI, 1.05-14.9). There was an increased IRR of lymphoma for males with Crohn disease only (3.63; 95% CI, 1.53-8.62), and this finding did not appear to be related to use of immunomodulatory therapy. Compared with controls, Crohn's disease was associated with an increased risk of cancer overall, but UC was not.

CONCLUSIONS

There appear to be similar increased risks for developing colon carcinoma and hepatobiliary carcinoma among patients with Crohn disease and UC. There is an increased risk of developing rectal carcinoma in UC patients, an increased risk of developing carcinoma of the small bowel in Crohn disease patients, and an increased risk of developing lymphoma among males with Crohn disease.

The incidence of fracture among patients with inflammatory bowel disease. A population-based cohort study

BACKGROUND

The clinical significance of the high prevalence of osteopenia in inflammatory bowel disease is unclear.

OBJECTIVE

To determine whether persons with inflammatory bowel disease have increased incidence of fracture.

DESIGN

Population-based matched cohort study.

SETTING

Manitoba, Canada.

PATIENTS

Patients with inflammatory bowel disease in the University of Manitoba IBD Database (n = 6027) were matched to 10 randomly selected persons in the general population without inflammatory bowel disease (n = 60 270) by year, age, sex, and postal area of residence.

MEASUREMENTS

The incidence of hospitalization for hip fracture was determined on the basis of hospital discharge abstracts. Outpatient medical billing records and hospital discharge abstracts were used to calculate the incidence of spine, rib, and forearm fractures. Rates were calculated on the basis of person-years of follow-up for 1984 to 1997.

RESULTS

Persons with inflammatory bowel disease had significantly increased incidence of fractures at the spine (incidence rate ratio [IRR], 1.74 [95% CI, 1.34 to 2.24]; P < 0.001), hip (IRR, 1.59 [CI, 1.27 to 2.00]; P < 0.001), wrist/forearm (IRR, 1.33 [CI, 1.11 to 1.58]; P = 0.001), and rib (IRR, 1.25 [CI, 1.02 to 1.52]; P = 0.03) and of any of these fractures (IRR, 1.41 [CI, 1.27 to 1.56]; P < 0.001).

CONCLUSION

The incidence of fracture among persons with inflammatory bowel disease is 40% greater than that in the general population.

Epidemiology of Crohn's disease and ulcerative colitis in a central Canadian province: a population-based study

The aim of this study was to assess the accuracy and utility of administrative health data in identifying persons with inflammatory bowel disease on a population basis and to determine the incidence and prevalence of this disease in the Canadian province of Manitoba. The data from Manitoba Health (the province's single insurer) were used to identify residents with physician and/or hospital contacts for Crohn's disease or ulcerative colitis based on International Classification of Diseases, Ninth Revision, Clinical Modification, codes between 1984 and 1995. Of 5,182 eligible individuals, 4,514 were mailed questionnaires and 2,725 responded. Cases were defined as individuals with five or more separate medical contacts with one of these diagnoses or three or more such contacts if they were resident for less than 2 years. The accuracy of the study case definitions was high when compared with either self-report or chart review. The 1989-1994 age- and sex-adjusted annual incidence was 14.6/100,000 for Crohn's disease and 14.3/100,000 for ulcerative colitis. The prevalence of Crohn's disease in 1994 was 198.5/100,000, and that of ulcerative colitis was 169.7/100,000. In conclusion, the authors have successfully established and validated a population-based database of inflammatory bowel disease based on administrative data. The high incidence rates and dynamic epidemiology of inflammatory bowel disease in Manitoba indicate the presence of important environmental risk factors, which warrants further investigation.

Incidence and prevalence of diabetes in children aged 0-14 years in Manitoba, Canada, 1985-1993

OBJECTIVE

To estimate the incidence and prevalence of type I diabetes among Manitoba children aged 0-14 years from 1985-1993.

RESEARCH DESIGN AND METHODS

The Manitoba Diabetes Database (a population-based database of individuals diagnosed with diabetes based on Manitoba's health insurance system) was used to estimate the annual incidence of diabetes for the years 1985-1993 and the point prevalence of diabetes at 31 March 1993 for Manitoba children aged 0-14 years. The Diabetes Education Resource for Children and Adolescents program database was used to correct incidence and prevalence rates for ascertainment using the two-source capture-recapture method.

RESULTS

The overall ascertainment rate of the Manitoba Diabetes Database was 95% for incident cases and 93% for prevalent cases. The average annual incidence was 20.4 per 100,000 for children aged 0-14 years. The annual incidence appears to be stable for all age-groups 0-14 years over the past decade. The point prevalence of diabetes among children was 120.4 per 100,000.

CONCLUSIONS

The incidence of type I diabetes in children aged 0-14 years in Manitoba is higher than reported previously in other urban regions of Canada, but similar to population-based estimates from Prince Edward Island. The incidence appears stable in Manitoba over the past decade even in the 0-4 year age-group. The Manitoba Diabetes Database appears to be a highly accurate population-based source of data on the epidemiology of diabetes in children.

Incidence and prevalence of diabetes in Manitoba, 1986-1991

OBJECTIVE

To estimate the incidence and prevalence of diabetes among adults in Manitoba, Canada, from 1986 to 1991.

RESEARCH DESIGN AND METHODS

A population-based database of individuals diagnosed with diabetes (Manitoba Diabetes Database) was created using data from Manitoba Health's comprehensive insurance system. Using this database, estimates of the annual incidence and prevalence of diabetes among Manitoba adults aged > or = 25 years were made for the years 1986-1991. Age-specific and age-adjusted rates were calculated separately for men and women.

RESULTS

The prevalence of diabetes in 1991 was 66.9/1,000 among adults > or = 25 years. Between 1986 and 1991 the age-adjusted prevalence rose steadily among both men and women. In 1991, the incidence of diabetes was 5.6/1,000. After an observed decline prior to 1989, the annual incidence of diabetes appears to be relatively stable.

CONCLUSIONS

The prevalence of diabetes is increasing steadily despite relatively stable incidence rates. Population-based data are required for projecting future trends and are an important tool for planning the required health resources.

Record linkage strategies, outpatient procedures, and administrative data

By understanding the range of approaches implicit in modern record linkage, epidemiologists and health services researchers can better decide its suitability for their needs. The authors discuss a small record linkage project, providing a sense of where mistakes were made. The research first uses existing identification numbers as a gold standard for linking hospital abstracts and physician claims to investigate whether or not coronary angiography was performed on a given individual. Even if identification numbers are not available, a successful linkage (with more than 95% of the cases matched) may be possible under some circumstances. The linkage process highlights problems with the consistent recording of coronary angiography in inpatient and outpatient hospital abstracts. Our approach should prove useful when the same procedure is recorded in more than one place on a single file and when validating a procedure (or other event) across files is important. Given the growing number of health care databases and ongoing changes in the delivery of care, record linkage often can provide quality control and expand research opportunities in a timely fashion.

Using administrative health data to monitor potential adverse health effects in environmental studies

Following episodes of environmental contamination, health professionals are limited in their ability to detect adverse health effects in surrounding communities due to lack of relevant baseline health data, resources, and appropriate control populations. The objective of this study was to ascertain the feasibility of using administrative health data for these purposes. The Manitoba Health Services Commission's (MHSC) database is comprehensive since universal health care is free in Canada. As part of an evaluation of two proposed hazardous waste treatment sites, the feasibility of using MHSC's data was tested by (a) defining the two study and control sites through use of MHSC's population registry and (b) determining baseline morbidity rates through analysis of MHSC's physician visit payment files; diagnoses were coded using ICD-9-CM. The results indicated that there were some differences between the groups studied in the age- and sex-standardized morbidity rates of diagnoses potentially influenced by exposures to chemicals. Use of administrative data provided by a national health service is an inexpensive and efficient way to create and follow potentially exposed cohorts residing in defined communities. Despite limitations related to small populations in exposed communities and lack of standardized diagnostic criteria by physicians, this method should be explored further in environmental studies.

Clinical effectiveness of influenza vaccination in Manitoba

OBJECTIVE

To assess the clinical effectiveness of influenza vaccination in preventing influenza-associated hospitalization and death.

DESIGN

Case-control study.

SETTING AND PATIENTS

Noninstitutionalized persons aged 45 years or older living in Manitoba, on December 1, 1982, and December 1, 1985.

METHODS

Linked records of the Manitoba population registry, hospital-discharge abstracts, physician claims for ambulatory-patient visits and influenza vaccination, and vital statistics were used. A matched-set analysis estimated the clinical effectiveness of influenza vaccination in preventing hospital admissions and deaths from influenza-associated conditions during influenza A (H3N2) outbreak periods in 1982 to 1983 (12 weeks) and 1985 to 1986 (10 weeks). The analysis adjusted for hospital discharge and ambulatory care for high-risk conditions within the previous 15 months and 3 months, respectively.

RESULTS

Influenza vaccination prevented 32% to 39% of hospital admissions with pneumonia and influenza and 15% to 34% of admissions with all respiratory conditions. Vaccination was 43% to 65% effective in preventing hospital deaths with these conditions (all listed diagnoses) and 27% to 30% effective in preventing deaths from all causes.

CONCLUSION

Influenza vaccination has substantial clinical effectiveness in preventing hospital admission and death from influenza-associated conditions in noninstitutionalized individuals.

Assessing physicians' compliance with guidelines for Papanicolaou testing

In this study, population-based data were used to examine the appropriateness of Papanicolaou (Pap) testing from the perspective of the women being tested and their physicians. The approach used is unique in its assessment of overtesting and undertesting in the primary care setting. From the data base of the province of Manitoba's universal health insurance plan, 4-year health histories (1981 to 1984) were constructed for each woman from a random sample of the population of women who, in 1982, were between the ages of 25 to 64 years (n = 22,287). At the last visit to a general practitioner, gynecologist, or general surgeon in 1984 (termed the current visit), the authors determined whether a Pap test was given for each woman. Using decision rules from a Canadian task force report on cervical screening and previous health history, the authors evaluated the appropriateness of screening by determining whether a Pap test was given and was needed, or whether a women who had not received a Pap test required one. Overall, 55.7% of women were tested appropriately. Of the 5352 women who received a Pap test at the current visit, 62.8% were overtested. Of the 16,935 women not tested at the current visit, 38.5% required screening (i.e. were undertested). Characteristics of a physician's practice that were significantly related to compliance with the guidelines included having a high proportion of patients visiting for obstetric or gynecologic reasons. Variables that were associated with negative compliance were 1) being a gynecologist; and 2) having a high proportion of patients who lived in inner city or rural areas. Because physicians are paid a fee for every Pap smear taken and the guidelines were well disseminated, these results should be reasonably representative of fee-for-service practice in North America, where preventive care is not subject to user charges. This study supports previous findings that a passive approach to dissemination of guidelines is insufficient to effect practice.

Disparity between influenza vaccination rates and risks for influenza-associated hospital discharge and death in Manitoba in 1982-1983

OBJECTIVE

To determine, in a defined population, the percentage of persons who were discharged from a hospital or died of influenza-associated respiratory conditions who had a health care contact during the preceding vaccination season and to determine the relation between risk status for influenza-associated hospitalization and death and influenza vaccination rates.

DESIGN

An observational study using linked-record analysis of medical claims data.

SETTING AND PATIENTS

A probability sample of 100,000 noninstitutionalized adults living in Manitoba in 1982 to 1983.

MEASUREMENTS

Analysis of medical claims for influenza vaccination and hospital discharges and deaths for influenza-associated respiratory conditions during the 1982-83 influenza vaccination season and influenza outbreak period.

RESULTS

For the population as a whole, 50% to 60% of elderly persons (greater than or equal to 65 years of age) and 30% to 40% of younger persons had one or more health care contacts during the influenza vaccination season but fewer than 10% of all persons had been discharged from a hospital. In contrast, for elderly persons hospitalized with respiratory conditions during the influenza outbreak period, approximately 80% had at least one health care contact during the vaccination season. Among the elderly, 39% to 46% of all those discharged for influenza-associated respiratory conditions and 62% to 67% of those who died had been discharged from hospital during the previous vaccination season. Persons discharged with high-risk conditions during the vaccination season were at greater risk for hospitalization with influenza-associated respiratory conditions but were less likely to be vaccinated than were those at lower risk.

CONCLUSIONS

Most persons who were hospitalized with influenza-associated respiratory conditions had contact with health care providers during the preceding influenza vaccination season. Among elderly patients, previous hospital care was common, especially among those who died. The disparity between influenza vaccination rates and risks for influenza-associated hospital discharge and death supports a strategy of hospital-based influenza vaccination.

Record linkage strategies: Part II. Portable software and deterministic matching

Software to perform record linkage should have several characteristics: (1) portability in being able to function with researchers' current arrangement of computer systems and languages, (2) flexibility in handling different linkage strategies, and (3) low cost in both computer time and researchers' efforts. A linkage package (LINKS) is described which satisfies these criteria; LINKS provides tools for both deterministic and probabilistic linkage as well as test modules for assessing data quality and structure. Because each linkage project is different, the modular nature of the software allows for better control of the programming process and development of unique strategies. Since the user provides the weights and decision rules, he may modify data between steps and/or develop extra steps to supplement the basic modules. In two information-rich linkage projects involving California AIDS data, LINKS identified mortality using deterministic approaches and permitted comparisons with other software and strategies. Flexible software and a deterministic approach would have eliminated the expensive key entry used to add full names and social security numbers as additional identifiers to one of the California data files.

The incidence of fracture of the proximal femur in two million Canadians from 1972 to 1984. Projections for Canada in the year 2006

Reported increases in the number of fractures of the proximal femur in Europe are greater than can be explained by demographic changes alone. This trend was assessed in Canada by examining hospital discharge records from the provinces of Saskatchewan and Manitoba from 1972 to 1984. The annual number of first fractures of the proximal femur in persons older than 50 years of age increased 59.7% in women and 42.2% in men during this time period. In most of the five-year age groups the percentage of increase in the number of fractures exceeded the percentage of increase in population of that age group. Annual age-specific incidences (by five-year age groups) increased exponentially with age, doubling every six years, and reached a maximum value of 4% in women older than 90 years of age. Annual age-adjusted incidences increased significantly over the study period in men and women. For the whole of Canada in 1987, it is estimated that there were 13,193 first fractures of the proximal femur in women and 4610 in men, and that in the year 2006 these will rise to 22,922 and 7846, respectively. The actual increase will be considerably greater if the age-specific incidences continue to increase as they have from 1972 to 1984. The gradual decline in physical activity, which contributes to bone loss, may be one etiological factor of this trend during the last half century.

Record linkage strategies. Part I: Estimating information and evaluating approaches

Record linkage techniques can help identify the same patient for matching diverse files (hospital discharge abstracts, insurance claims, registries, Vital Statistics data) which contain similar identifiers. Prior knowledge of whether a linkage is feasible is important to prevent wasted effort (additional data collection or data manipulation), which decreases the cost-effectiveness of the linkage. Using examples generated by linking the Manitoba Health Services Commission data with Vital Statistic files, a method of estimating the information in each data set is presented first. Further, the feasibility of several different record linkage strategies is described and tested, given varying amounts of information. At the margin, relatively small amounts of information (having just one more variable to match with) can make a great difference. Probabilistic linkage's great advantage was found in those situations where only a moderate amount of extra information was available. By using the above techniques when working with one or both files in a proposed record linkage project, a much more informed judgement can now be made as to whether a linkage will or will not work. In facilitating record linkage, flexibility of both software and the strategy for matching is very important.

Risk adjustment in claims-based research: the search for efficient approaches

Claims-based indices of comorbidity and severity, as well as other measures derived from routinely collected administrative data, are developed and tested. The extent to which risk adjustments using claims can be improved by adding information from one well-known measure based on chart review and patient examination (the American Society of Anesthesiologists' (ASA) Physical Status score) is also examined. Readmissions and mortality after three common surgical procedures are the outcomes studied using multiple logistic regression. Claims-based measures of comorbidity, derived both from hospital discharge abstracts at the time of surgery and from hospitalizations in the 6 months before surgery, provided reasonably good predictions of postsurgical readmissions and mortality. In the most complete logistic regression models, the Somers' Dyx measure of fit (a rank correlation coefficient) ranged from 0.23 to 0.38 for readmissions and from 0.46 to 0.72 for mortality. In 5 out of 6 cases, these predictions were not improved by including the prospectively-collected ASA Physical Status score. Such difficulties in improving risk adjustment by more intensive data collection are discussed in terms of their research implications.

Assessing data quality: a computerized approach

With the growing reliance on large health care data bases, the need to verify data quality increases as well. Because of the considerable costs involved in checks using primary data collection, a computerized methodology for performing such checks is suggested. The technique seems appropriate for any situation where two data collection systems (i.e. hospital discharge abstracts and physician claims for payment) relate to the same event, such as a patient's hospitalization. After reviewing other approaches, this paper suggests linking physician claims for performing particular surgical procedures with hospital discharge abstracts for the stay in which the surgery took place. Physician and hospital data for adults age 25 and over in Manitoba from 1 April, 1979 to 31 March, 1984 were used to address the questions: 1. How well can the two data sets be linked? 2. Given linkage of the two data sets, how much agreement is there as to procedure and diagnosis? Linkage between hospital and physician data was excellent (over 95%) for 5 out of 11 surgical procedures (hysterectomy, prostatectomy, total hip replacement, coronary artery bypass surgery, and heart valve replacement); there was over 90% perfect agreement for three other procedures (cholecystectomy, cataract surgery and total knee replacement). Problems with matching the Manitoba Health Services Commission tariffs (on physician claims) with ICD-9-CM operation codes (on hospital data) led to only 77% perfect agreement for vascular surgery and 84% for gallbladder and biliary tract operations other than cholecystectomy; over 10% of the cases linked on surgeon and date but not on the designated procedures.(ABSTRACT TRUNCATED AT 250 WORDS)

Is there a role for surgery in the acute management of infective endocarditis? A decision analysis and medical claims database approach

In the absence of good clinical evidence from a randomized trial, the authors performed a decision analysis to determine the potential value of early elective surgery (OPNOW) for patients with left-sided Staphylococcus aureus infective endocarditis. Initial impressions (before performance of decision analysis) and initial runs at the formal models using probability estimates derived from clinicians suggested that OPNOW (i.e., within a few days of starting antibiotics) offered no advantage over attempted medical cure (WAIT) (life expectancy: WAIT = 325 weeks; OPNOW = 255 weeks). Extensive sensitivity analyses identified critical variables that needed further empirical estimation. The Manitoba Health Services Commission database identified 127 incident cases of endocarditis between April 1, 1979, and March 31, 1985, enabling estimation of values for these critical variables. With these estimates, the early surgery strategy appeared much better than the previous analyses had suggested (life expectancy: WAIT = 208 weeks, OPNOW = 256 weeks). The authors believe that this approach of combining decision analysis with medical claims databases is useful as an alternative or precursor to randomized trials, especially where the resource requirements and logistic difficulties of performing randomized trials are great.

Software for health care analysts: a modular approach

The many concerns about the cost and quality of health care suggest the need to facilitate planners' using existing data bases for utilization review, program evaluation, and technology assessment. Despite both the availability of relevant data and widespread improvements in computing power, integrated computer software to permit analyses by nonspecialists has not previously been developed. This paper discusses the features of a health policy information system which aids working with hospital discharge abstracts, medical claims, cancer registries, and vital statistics files. Analyses of small area utilization, length of stay, in-hospital mortality, and readmissions are facilitated by this package. This information system, named the Health Applications System, includes an analysis module, three information management modules, and a set of record linkage modules. The modules were developed using the macroprocessor in the fourth-generation SAS system. Features of the software and their implications for data analysis are discussed.

Simplifying record linkage: software and strategy

Although the methodology of record linkage is fairly well developed, there is a need for less expensive methods and simpler software to facilitate trying out different tactics to generate good linkages. The present work has built on a fourth generation language SAS (Statistical Analysis System) with accompanying macroprocessor, to develop a user-friendly and flexible system for both exact and probabilistic matching. The major features of the LINKS system are presented and illustrated using 1979-1984 information from the Manitoba Health Services Commission (MHSC) registry file with the Canadian Mortality Data Base. Initial runs with exact, then probabilistic, matching linked approximately 91% of the Vital Statistics records to corresponding MHSC records. Subsequent modification of parameters improved the linkage to 95%.

The art and science of record linkage: methods that work with few identifiers

Increasing use of administrative records for research purposes has led to the need to develop methods for linking records. Linkage of the data from the Manitoba Health Services Commission (MHSC) with Canadian Vital Statistics records was done to verify Manitoba deaths, to measure degree of agreement between the two data sets, and to link "cause of death" information from Vital Statistics back to medical histories. A multi-stage approach to matching was adopted. Individual records matching perfectly across a number of relevant variables in the two files were selected first, reducing the number of records going through the probability matching to manageable proportions. Overall, almost 96% of the MHSC records were well matched with Vital Statistics records. Both uses of the matched data and various ways to confirm data quality are discussed.

Variations in physicians' hospitalization practices: a population-based study in Manitoba, Canada

This paper uses claims data from a universal health care system to describe physicians' hospitalization styles after adjusting for case-mix characteristics of their primary patients. Patients were uniquely assigned to that physician (general or family practitioners, internist, general surgeon, or obstetrician/gynecologist) seen most frequently over each two two-year periods (1972-74 and 1974-76). Four indices were developed including: 1) percentage of primary patients hospitalized; 2) mean number of readmissions for such patients; 3) mean length of stay; and 4) total days of hospitalization per primary care patient (a summary measure combining the first three). Rates of admission, not length of stay, were shown to be strongly related to this summary measure. Marked variations in the hospitalization indices were observed across physicians; these variations cannot be explained by the health or sociodemographic characteristics of a physician's patients. Rural physicians practicing in areas with high bed-to-population ratios and low occupancy rates were particularly high users of hospitals. The economic implications of different practice styles are shown to be large; physicians who were high users of hospitals serve 27 per cent of the patients but their patients consume 42 per cent of the hospital days.