Finding incident breast cancer cases through US claims data and a state cancer registry

Finding incident cancer cases through outpatient oncology clinic claims data and integration into a state cancer registry

Cancer data from population-based cancer registries under-report cancer cases, especially for cancers primarily diagnosed and treated in outpatient clinical settings, away from hospital-based cancer registrars. Previously, we developed alternative methods of cancer case capture including a claims-based method, which identified a large proportion of cancer cases missed by traditional population-based cancer registries. In this study, we adapted a claims-based method for statewide implementation of cancer surveillance in Florida. Between 2010 and 2017 the claims-based method identified 143,083 cancer abstracts, of which 42% were new and 58% were previously registered. The claims-based method led to the creation of 53,419 new cancer cases in the state cancer registry, which made up 9.3% of all cancer cases registered between 2010 and 2017. The types of cancers identified by the claims-based method were typical of the kinds primarily diagnosed and treated in outpatient oncology clinic settings, such as hematological malignancies, prostate cancer, melanoma, breast cancer, and bladder cancer. These cases were added to the Florida cancer registry and may produce an artefactual increase in cancer incidence, which is believed to be closer to the actual burden of cancer in the state.

Leveraging Linkage of Cohort Studies With Administrative Claims Data to Identify Individuals With Cancer

BACKGROUND

In an effort to overcome quality and cost constraints inherent in population-based research, diverse data sources are increasingly being combined. In this paper, we describe the performance of a Medicare claims-based incident cancer identification algorithm in comparison with observational cohort data from the Nurses' Health Study (NHS).

METHODS

NHS-Medicare linked participants' claims data were analyzed using 4 versions of a cancer identification algorithm across 3 cancer sites (breast, colorectal, and lung). The algorithms evaluated included an update of the original Setoguchi algorithm, and 3 other versions that differed in the data used for prevalent cancer exclusions.

RESULTS

The algorithm that yielded the highest positive predictive value (PPV) (0.52-0.82) and κ statistic (0.62-0.87) in identifying incident cancer cases utilized both Medicare claims and observational cohort data (NHS) to remove prevalent cases. The algorithm that only used NHS data to inform the removal of prevalent cancer cases performed nearly equivalently in statistical performance (PPV, 0.50-0.79; κ, 0.61-0.85), whereas the version that used only claims to inform the removal of prevalent cancer cases performed substantially worse (PPV, 0.42-0.60; κ, 0.54-0.70), in comparison with the dual data source-informed algorithm.

CONCLUSIONS

Our findings suggest claims-based algorithms identify incident cancer with variable reliability when measured against an observational cohort study reference standard. Self-reported baseline information available in cohort studies is more effective in removing prevalent cancer cases than are claims data algorithms. Use of claims-based algorithms should be tailored to the research question at hand and the nature of available observational cohort data.

Obstetrical and infant outcomes among women with neoplasms during pregnancy

PURPOSE

One in 1,000 pregnancies is complicated by malignancies. Prevalence is greater for benign neoplasms. Adverse outcomes among women with malignancies have been reported. Less is known of postpartum outcomes for infants, or outcomes among women with benign neoplasms.

METHODS

We conducted a population-based cohort study using Washington State-linked vital-hospital discharge records. Women with neoplasms (707 malignant; 13,156 benign) with deliveries in 1987-2012 were identified, and a randomly selected comparison cohort. Obstetrical/infant outcomes and rehospitalization < 2 years post-delivery were compared separately for each group by multivariable regressions to estimate risk ratios (RR) and 95% confidence intervals (CI).

RESULTS

Women with either condition had increased anemia, cesarean, and preterm delivery; their infants were more often < 2,500 g or jaundiced. Women with benign conditions had increased gestational diabetes (RR = 1.20; 95% CI 1.12-1.28) and preeclampsia (RR = 1.27; 95% CI 1.18-1.36); their infants had increased malformations (RR = 1.29; 95% CI 1.19-1.38). Women with neoplasms more often were hospitalized seven or more days or rehospitalized; their infants' hospitalizations were also longer.

CONCLUSION

Malignant and benign neoplasms were associated with several adverse outcomes. Reasons for relationships of benign neoplasms with gestational diabetes, preeclampsia, and congenital malformations merit further study.

Accuracy of administrative databases in detecting primary breast cancer diagnoses: a systematic review

OBJECTIVE

To define the accuracy of administrative datasets to identify primary diagnoses of breast cancer based on the International Classification of Diseases (ICD) 9th or 10th revision codes.

DESIGN

Systematic review.

DATA SOURCES

MEDLINE, EMBASE, Web of Science and the Cochrane Library (April 2017).

ELIGIBILITY CRITERIA

The inclusion criteria were: (a) the presence of a reference standard; (b) the presence of at least one accuracy test measure (eg, sensitivity) and (c) the use of an administrative database.

DATA EXTRACTION

Eligible studies were selected and data extracted independently by two reviewers; quality was assessed using the Standards for Reporting of Diagnostic accuracy criteria.

DATA ANALYSIS

Extracted data were synthesised using a narrative approach.

RESULTS

From 2929 records screened 21 studies were included (data collection period between 1977 and 2011). Eighteen studies evaluated ICD-9 codes (11 of which assessed both invasive breast cancer (code 174.x) and carcinoma in situ (ICD-9 233.0)); three studies evaluated invasive breast cancer-related ICD-10 codes. All studies except one considered incident cases.The initial algorithm results were: sensitivity ≥80% in 11 of 17 studies (range 57%-99%); positive predictive value was ≥83% in 14 of 19 studies (range 15%-98%) and specificity ≥98% in 8 studies. The combination of the breast cancer diagnosis with surgical procedures, chemoradiation or radiation therapy, outpatient data or physician claim may enhance the accuracy of the algorithms in some but not all circumstances. Accuracy for breast cancer based on outpatient or physician's data only or breast cancer diagnosis in secondary position diagnosis resulted low.

CONCLUSION

Based on the retrieved evidence, administrative databases can be employed to identify primary breast cancer. The best algorithm suggested is ICD-9 or ICD-10 codes located in primary position.

TRIAL REGISTRATION NUMBER

CRD42015026881.

Evaluation of algorithms to identify incident cancer cases by using French health administrative databases

PURPOSE

Administrative databases are increasingly being used in cancer observational studies. Identifying incident cancer in these databases is crucial. This study aimed to develop algorithms to estimate cancer incidence by using health administrative databases and to examine the accuracy of the algorithms in terms of national cancer incidence rates estimated from registries.

METHODS

We identified a cohort of 463 033 participants on 1 January 2012 in the Echantillon Généraliste des Bénéficiaires (EGB; a representative sample of the French healthcare insurance system). The EGB contains data on long-term chronic disease (LTD) status, reimbursed outpatient treatments and procedures, and hospitalizations (including discharge diagnoses, and costly medical procedures and drugs). After excluding cases of prevalent cancer, we applied 15 algorithms to estimate the cancer incidence rates separately for men and women in 2012 and compared them to the national cancer incidence rates estimated from French registries by indirect age and sex standardization.

RESULTS

The most accurate algorithm for men combined information from LTD status, outpatient anticancer drugs, radiotherapy sessions and primary or related discharge diagnosis of cancer, although it underestimated the cancer incidence (standardized incidence ratio (SIR) 0.85 [0.80-0.90]). For women, the best algorithm used the same definition of the algorithm for men but restricted hospital discharge to only primary or related diagnosis with an additional inpatient procedure or drug reimbursement related to cancer and gave comparable estimates to those from registries (SIR 1.00 [0.94-1.06]).

CONCLUSION

The algorithms proposed could be used for cancer incidence monitoring and for future etiological cancer studies involving French healthcare databases. Copyright © 2017 John Wiley & Sons, Ltd.

Estimation of cancer cases using capture-recapture method in Northwest Iran

BACKGROUND

Under-ascertainment and over-ascertainment are common phenomena in surveillance and registry systems of health-related events. Capture-recapture is one of the methods which is applied to determine the sensitivity of surveillance or registry systems to recognize cancer cases. This study aimed to estimate the number of cancers using data available both in the Cancer Registry Center of Northwestern Iran and in the Population-based Cancer Registry Center of Iran.

MATERIAL AND METHODS

The studied population consisted of all cancerous cases in the northwest of Iran from 2008 to 2010. All data were extracted from two resources and entered into Microsoft Excel software. After removing common and repeat cases the data were statistically analyzed using a capture-recapture studies' specific software "CARE 1.4". Estimations were calculated by Chapman and Petersen methods with the approximate confidence interval of 95%.

RESULTS

From 2008 to 2010, the number of all cancer cases was estimated to be 21,652 (CI 95%: 19,863-22,101). Sensitivity rate of all cancer cases was 83.9% and that of Population-based Cancer Registry Center of Iran was 52%. It was 93.1% considering both resources.

CONCLUSION

Using two resources and the capture-recapture method rather than a single resource may be a more reliable method to estimate the number of cancer cases.

Using hospital discharge data to identify incident pregnancy-associated cancers: a validation study

BACKGROUND

Pregnancy-associated cancer is associated with maternal morbidities and adverse pregnancy outcomes, and is reported to be increasing. Hospital discharge data have the potential to provide timely information on cancer incidence, which is central to evaluation and improvement of clinical care for women. This study aimed to assess the validity of hospital data for identifying incident pregnancy-associated cancers compared with incident cancers from an Australian population-based statutory cancer registry.

METHODS

Birth data from 2001-2008, comprised 470,277 women with 679,736 maternities, were linked to cancer registry and hospitalisation records to identify newly diagnosed cancers during pregnancy or within 12 months of delivery. Two hospital-identified cancer groups were examined; "index cancer hospitalisation" - first cancer admission per woman per pregnancy and "all cancer hospitalisations" -the total number of hospitalisations with a cancer diagnosis and women could have multiple hospitalisations during pregnancy. The latter replicates a scenario where identification of individuals is not possible and hospitalisations are used as the unit of analysis.

RESULTS

The incidence of pregnancy-associated cancer (according to cancer registry) was 145.4/100,000 maternities. Incidence of cancer was substantially over-estimated when using hospitalisations as the unit of analysis (incidence rate ratio, IRR 1.7) and under-estimated when using the individual (IRR 0.8). Overall, the sensitivity of "index cancer hospitalisation" was 60.4%, positive predictive value (PPV) 77.7%, specificity and negative predictive value both 100%. Melanoma ascertainment was only 36.1% and breast cancer 62.9%. For other common cancers sensitivities ranged from 72.1% to 78.6% and PPVs 56.4% to 87.3%.

CONCLUSION

Although hospital data provide another timely source of cancer identification, the validity is insufficient to obtain cancer incidence estimates for the obstetric population.

Estimation of asthma incidence among low-income children in Texas: a novel approach using Medicaid claims data

Few recent estimates of childhood asthma incidence exist in the literature, although the importance of incidence surveillance for understanding asthma risk factors has been recognized. Asthma prevalence, morbidity, and mortality reports have repeatedly shown that low-income children are disproportionately impacted by the disease. The aim of this study was to demonstrate the utility of Medicaid claims data for providing statewide estimates of asthma incidence. Medicaid analytic extract (MAX) data for Texas children aged 0-17 years enrolled in Medicaid between 2004 and 2007 were used to estimate incidence overall and by age group, gender, race, and county of residence. A ≥13-month period of continuous enrollment was required in order to distinguish incident from prevalent cases identified in the claims data. The age-adjusted incidence of asthma was 4.26/100 person-years during 2005-2007, higher than reported in other populations. Incidence rates decreased with age, were higher for males than females, differed by race, and tended to be higher in rural than urban areas. This study demonstrates the utility of Medicaid analytic extract data for estimating asthma incidence and describes the methodology required for a population with unstable enrollment.

Hospital discharge diagnostic and procedure codes for upper gastro-intestinal cancer: how accurate are they?

BACKGROUND

Population-level health administrative datasets such as hospital discharge data are used increasingly to evaluate health services and outcomes of care. However information about the accuracy of Australian discharge data in identifying cancer, associated procedures and comorbidity is limited. The Admitted Patients Data Collection (APDC) is a census of inpatient hospital discharges in the state of New South Wales (NSW). Our aim was to assess the accuracy of the APDC in identifying upper gastro-intestinal (upper GI) cancer cases, procedures for associated curative resection and comorbidities at the time of admission compared to data abstracted from medical records (the 'gold standard').

METHODS

We reviewed the medical records of 240 patients with an incident upper GI cancer diagnosis derived from a clinical database in one NSW area health service from July 2006 to June 2007. Extracted case record data was matched to APDC discharge data to determine sensitivity, positive predictive value (PPV) and agreement between the two data sources (κ-coefficient).

RESULTS

The accuracy of the APDC diagnostic codes in identifying site-specific incident cancer ranged from 80-95% sensitivity. This was comparable to the accuracy of APDC procedure codes in identifying curative resection for upper GI cancer. PPV ranged from 42-80% for cancer diagnosis and 56-93% for curative surgery. Agreement between the data sources was >0.72 for most cancer diagnoses and curative resections. However, APDC discharge data was less accurate in reporting common comorbidities - for each condition, sensitivity ranged from 9-70%, whilst agreement ranged from κ = 0.64 for diabetes down to κ < 0.01 for gastro-oesophageal reflux disorder.

CONCLUSIONS

Identifying incident cases of upper GI cancer and curative resection from hospital administrative data is satisfactory but under-ascertained. Linkage of multiple population-health datasets is advisable to maximise case ascertainment and minimise false-positives. Consideration must be given when utilising hospital discharge data alone for generating comorbidity indices, as disease burden at the time of admission is under-reported.

Improving primary care for older adults with cancer and depression

BACKGROUND

Depression is common among older cancer patients, but little is known about the optimal approach to caring for this population. This analysis evaluates the effectiveness of the Improving Mood-Promoting Access to Collaborative Treatment (IMPACT) program, a stepped care management program for depression in primary care patients who had an ICD-9 cancer diagnosis.

METHODS

Two hundred fifteen cancer patients were identified from the 1,801 participants in the parent study. Subjects were 60 years or older with major depression (18%), dysthymic disorder (33%), or both (49%), recruited from 18 primary care clinics belonging to 8 health-care organizations in 5 states. Patients were randomly assigned to the IMPACT intervention (n = 112) or usual care (n = 103). Intervention patients had access for up to 12 months to a depression care manager who was supervised by a psychiatrist and a primary care provider and who offered education, care management, support of antidepressant management, and brief, structured psychosocial interventions including behavioral activation and problem-solving treatment.

RESULTS

At 6 and 12 months, 55% and 39% of intervention patients had a 50% or greater reduction in depressive symptoms (SCL-20) from baseline compared to 34% and 20% of usual care participants (P = 0.003 and P = 0.029). Intervention patients also experienced greater remission rates (P = 0.031), more depression-free days (P < 0.001), less functional impairment (P = 0.011), and greater quality of life (P = 0.039) at 12 months than usual care participants.

CONCLUSIONS

The IMPACT collaborative care program appears to be feasible and effective for depression among older cancer patients in diverse primary care settings.

Validating regional Hospital Information System data through comparison with a local cancer register to identify interval cancers of a breast screening program

The objective of this study was to evaluate the accuracy of the Hospital Information System (HIS) in monitoring the breast cancer incidence and interval cancers compared with the cancer registry (CR). The HIS data linked with CR and Mammographic Screening Information System data for breast cancer cases diagnosed in the period 1999-2003. The sensitivity and positive predictive value of the HIS data were calculated using the CR as a gold standard. One thousand two hundred and thirty-six breast cancers were registered by the CR and 1028 were reported in the HIS. The sensitivity rate was 83.2% and the positive predictive value was 83.0%; similar results were obtained in the screening target population (50-69 years old). Fifteen invasive breast cancers occurred among screened women identified by HIS (four interval cancers and 11 screen-detected), two were registered as in situ in the CR. The HIS seems to have the potential to identify interval breast cancers, but the low accuracy of information does not permit an exact measure of the incidence.

Is it possible to estimate the incidence of breast cancer from medico-administrative databases?

One approach to estimate cancer incidence in the French Départements is to quantify the relationship between data in cancer registries and data obtained from the PMSI (Programme de Médicalisation des Systèmes d'Information Médicale). This relationship may then be used in Départements without registries to infer the incidence from local PMSI data. We present here some methodological solutions to apply this approach. Data on invasive breast cancer for 2002 were obtained from 12 Départemental registries. The number of hospital stays was obtained from the National PMSI using two different algorithms based on the main diagnosis only (Algorithm 1) or on that diagnosis associated to a mention of "resection" (Algorithm 2). Considering registry data as gold standard, a calibration approach was used to model the ratio of the number of hospital stays to the number of incident cases. In Départements with registries, validation of the predictions was done through cross-validation. In Départements without registries, validation was done through a study of homogeneity of the mean number of hospital stays per patient. Cross-validation showed that the estimates predicted by the model were true with data extracted by Algorithm 1 but not by Algorithm 2. However, with Algorithm 1, there was an important heterogeneity between French Départements as to the mean number of hospital stays per patient, which had an important impact on the estimations. In the near future, the method will allow using medico-administrative data (after calibration with registry data) to estimate Départemental incidence of selected cancers.

A high positive predictive value algorithm using hospital administrative data identified incident cancer cases

OBJECTIVE

We have developed and validated an algorithm based on Piedmont hospital discharge abstracts for ascertainment of incident cases of breast, colorectal, and lung cancer.

STUDY DESIGN AND SETTING

The algorithm training and validation sets were based on data from 2000 and 2001, respectively. The validation was carried out at an individual level by linkage of cases identified by the algorithm with cases in the Piedmont Cancer Registry diagnosed in 2001.

RESULTS

The sensitivity of the algorithm was higher for lung cancer (80.8%) than for breast (76.7%) and colorectal (72.4%) cancers. The positive predictive values were 78.7%, 87.9%, and 92.6% for lung, colorectal, and breast cancer, respectively. The high values for colorectal and breast cancers were due to the model's ability to distinguish prevalent from incident cases and to the accuracy of surgery claims for case identification.

CONCLUSIONS

Given its moderate sensitivity, this algorithm is not intended to replace cancer registration, but it is a valuable tool to investigate other aspects of cancer surveillance. This method provides a valid study base for timely monitoring cancer practice and related outcomes, geographic and temporal variations, and costs.

Agreement of diagnosis and its date for hematologic malignancies and solid tumors between medicare claims and cancer registry data

PURPOSE

Claims data may be a suitable source studying associations between drugs and cancer. However, linkage between cancer registry and claims data including pharmacy-dispensing information is not always available. We examined the accuracy of claims-based definitions of incident cancers and their date of diagnosis.

METHODS

Four claims-based definitions were developed to identify incident leukemia, lymphoma, lung, colorectal, stomach, and breast cancer. We identified a cohort of subjects aged >or=65 (1997-2000) from Pennsylvania Medicare and drug benefit program data linked with the state cancer registry. We calculated sensitivity, specificity, and positive predictive values of the claims-based definitions using registry as the gold standard. We further assessed the agreement between diagnosis dates from two data sources.

RESULTS

All definitions had very high specificity (>or=98%), while sensitivity varied between 40% and 90%. Test characteristics did not vary systematically by age groups. The date of first diagnosis according to Medicare data tended to be later than the date recorded in the registry data except for breast cancer. The differences in dates of first diagnosis were within 14 days for 75% to 88% of the cases. Bias due to outcome misclassification of our claims-based definition of cancer was minimal in our example of a cohort study.

CONCLUSIONS

Claims data can identify incident hematologic malignancies and solid tumors with very high specificity with sufficient agreement in the date of first diagnosis. The impact of bias due to outcome misclassification and thus the usefulness of claims-based cancer definitions as cancer outcome markers in etiologic studies need to be assessed for each study setting.

French claims data as a source of information to describe cancer incidence: predictive values of two identification methods of incident prostate cancers

Claims data from the "Programme de Médicalisation du Système d'Information" (PMSI) have been commonly used for several years to complement cancer registries and describe cancer incidence in France. It is less clear whether or not it is possible to use these data as an independent source of information to assess cancer incidence, in the absence of a regional cancer registry. Following a similar study on breast cancer, we present a study which aimed to evaluate two methods of identifying incident prostate cancer using claims data. These methods were developed using claims data from the Hospices Civils de Lyon (HCL) and their validity was tested against medical records. The first method (M1) identified incident patients as those who had at least one stay with a principal diagnosis of prostate cancer. The second method (M2) had a prostate cancer treatment code in addition to the criteria for the first method. Both methods of identification had similar results, indicating a low rate of false negatives (negative predictive values: M1 = 100 [CI95: 93.8-100], M2 = 98.6 [CI95: 90.1-99.6]) and a high rate of false positives (positive predictive values: M1 = 33.3 [CI95: 23.2-42.1], M2 = 33.7 [CI95: 24.2-43.2]). The sample size did not allow us to produce consistent estimates of sensitivity and specificity. Our results showed that an estimation of the number of incident cases of prostate cancer using both methods of identification would be biased because of the high rate of false positives. Statistical methods that correct identification errors should be used.

[Comparison of the French DRG based information system (PMSI) with the National Mesothelioma Surveillance Program database]

BACKGROUND

One of the main purposes of the French National Mesothelioma Surveillance Program is to estimate and follow the national incidence of pleural mesothelioma. We wanted to study the contribution of the French hospital national database as a valid source of mesothelioma incident cases.

METHODS

From the 1998 and 1999 hospital national database, medical records with a diagnosis code of mesothelioma or pleural cancer where selected among patients who resided in one of the 17 administrative divisions covered by the National Mesothelioma Surveillance Program in 1998. From these records, 506 patients in 1998 and 474 patients in 1999 where identified and matched with the National Mesothelioma Surveillance Program cases over the same period using indirect criteria of identification (sex, age, place of residence). Medical records of cases unknown by the National Mesothelioma Surveillance Program where consulted in one of the administrative divisions.

RESULTS

Only two-thirds of the registered cases of the National Mesothelioma Surveillance Program could be matched with a patient identified in the hospital national database with a diagnosis of mesothelioma registered during the same year. Consultation of the medical records showed that 1) certified cases registered in the National Mesothelioma Surveillance Program where often (83%) found in the hospital national database with a code of mesothelioma but 10 to 15% of the patients with a code of mesothelioma in the national hospital databases had a different diagnosis according to their medical records; 2) 65% of the patients with a code of mesothelioma in the national hospital databases that where unknown from the National Mesothelioma Surveillance Program in 1998 and 55% in 1999 where prevalent cases; 3) 3 suspected cases had not been reported to the National Mesothelioma Surveillance Program.

CONCLUSION

Because of lack of diagnosis certification, mistakes in encoding diagnosis and the fact that incident and prevalent cases cannot be distinguished in the hospital national database make it impossible to estimate the mesothelioma incidence solely from this source of data. However, the hospital claim databases constitute a complementary source of information for the active search of incident cases performed by the National Mesothelioma Surveillance Program.

In the absence of cancer registry data, is it sensible to assess incidence using hospital separation records?

BACKGROUND

Within the health literature, a major goal is to understand distribution of service utilisation by social location. Given equivalent access, differential incidence leads to an expectation of differential service utilisation. Cancer incidence is differentially distributed with respect to socioeconomic status. However, not all jurisdictions have incidence registries, and not all registries allow linkage with utilisation records. The British Columbia Linked Health Data resource allows such linkage. Consequently, we examine whether, in the absence of registry data, first hospitalisation can act as a proxy measure for incidence, and therefore as a measure of need for service.

METHODS

Data are drawn from the British Columbia Linked Health Data resource, and represent 100% of Vancouver Island Health Authority cancer registry and hospital records, 1990-1999. Hospital separations (discharges) with principal diagnosis ICD-9 codes 140-208 are included, as are registry records with ICDO-2 codes C00-C97. Non-melanoma skin cancer (173/C44) is excluded. Lung, colorectal, female breast, and prostate cancers are examined separately. We compare registry and hospital annual counts and age-sex distributions, and whether the same individuals are represented in both datasets. Sensitivity, specificity and predictive values are calculated, as is the kappa statistic for agreement. The registry is designated the gold standard.

RESULTS

For all cancers combined, first hospitalisation counts consistently overestimate registry incidence counts. From 1995-1999, there is no significant difference between registry and hospital counts for lung and colorectal cancer (p = 0.42 and p = 0.56, respectively). Age-sex distribution does not differ for colorectal cancer. Ten-year period sensitivity ranges from 73.0% for prostate cancer to 84.2% for colorectal cancer; ten-year positive predictive values range from 89.5% for female breast cancer to 79.35% for prostate cancer. Kappa values are consistently high.

CONCLUSION

Claims and registry databases overlap with an appreciable proportion of the same individuals. First hospital separation may be considered a proxy for incidence with reference to colorectal cancer since 1995. However, to examine equity across cancer health services utilisation, it is optimal to have access to both hospital and registry files.

The added value of claims for cancer surveillance: results of varying case definitions

OBJECTIVE

As cancer diagnosis and treatment has moved to the outpatient healthcare setting, traditional cancer surveillance tools are less effective for complete and unbiased capture of incident cases. This study evaluates the potential for Medicare data to supplement cancer surveillance in a unique manner by using a standard that is independent of a central cancer registry.

DESIGN

State cancer registry records were matched with Medicare data. Case validation included inpatient record abstraction combined with a mail/telephone survey of treating physicians. The positive predictive value (PPV), sensitivity (capture rate), and potential additional cases were calculated for 6 Medicare claims-based case definitions.

RESULTS

The PPV varied according to cancer site and definition, ranging from 70%-97% (prostate) to 87%-98% (breast). Sensitivity varied inversely with PPV, ranging from 51%-94% (breast) to 10%-88% (lung). The most important factors that predicted being missed by the registry were having no admission to an ACOS-certified hospital and no surgical treatment.

CONCLUSION

Medicare data represent a valid resource for supplementing state cancer registries in surveillance efforts. This potential is especially applicable to cancers predominantly diagnosed and treated outside the hospital setting.

[Sensitivity and specificity of two methods used to identify incident breast cancer in specialized units using claims databases]

BACKGROUND

Hospital claims databases from acute care units are available nationwide and contain most patients at the beginning of their cancer. The goal is to define the ability of these databases to provide a number of incident breast cancer cases using identification methods. Two identification methods were assessed in three specialized sections of a teaching hospital.

METHODS

The first method identified women who had at least one stay with a principal diagnosis of breast cancer. The second, which is more restrictive, identified women who had at least one stay with a principal diagnosis of breast cancer and a breast cancer-specific surgical treatment code. Both methods were applied to 4588 women 20 Years of age or older hospitalized in three specialized sections of the Hospices Civils de Lyon in 2000. To categorize these women in two groups, incident breast cancer cases or non-incident breast cancer cases, 150 women were randomized in each of two groups, one for incident breast cancer cases and one for non-incident breast cancer cases. Their medical records were used as references.

RESULTS

Sensitivity, specificity and their credibility intervals were respectively 99.4% (84-99.9) and 91.7% (90.3-93.3) for the first method and 93.8% (76.2-98.7) and 97.3% (96.1-98) for the second. Among women wrongly identified with an incident breast cancer in 2000, 75.4% (43/57) had a breast cancer that was not incident that Year with the first method, compared to 96% (24/25) with the second. Among these women wrongly identified with an incident breast cancer, coding errors of the principal diagnosis were found for 24.6% (14/57) of patients with the first method and for 4% (1/25) with the second. Their correction led to 99.2% (86.5-99.9) sensitivity and 92.9% (91.4-94.6) specificity for the first method and to 94.2% (76.5-98.7) sensitivity and 97.3% (96.2-98.1) specificity for the second.

CONCLUSIONS

The second method using cancer-specific surgical codes appeared more specific with a slight loss in sensitivity. The use of identification methods to assess the number of incident cancer cases still have to be defined.

Using Medicare Data to Estimate the Number of Cases Missed by a Cancer Registry

BACKGROUND

Cancer surveillance is essential for assessing patterns of cancer occurrence. State cancer registries do not capture all available cases potentially biasing results. Secondary data may be useful in identifying new cases and estimating the number of cases missed.

OBJECTIVE

We sought to create 2 distinct data sources from Medicare claims to use in combination with registry data as 3 sources for a capture-recapture analysis to estimate the capture rate and bias in capture of a statewide cancer registry.

METHODS

Data from the Virginia cancer registry (Registry) were merged with Medicare inpatient (Part A) as well as Medicare outpatient and physician claims (Part B) to provide 3 sources to estimate missing cases. A 3-source loglinear model was used to estimate the number of missing cancer cases for breast, lung, colorectal, and prostate cancer. Models included main effects and interactions. Additional analysis looked at the effect of demographic and comorbidity variables.

RESULTS

Loglinear models demonstrated mostly positive dependence between the 3 sources, implying that 2-source models would underestimate missing cases and overestimate capture rates. Using capture-recapture estimates of total number of cancer cases as the denominator, capture rates for Registry ranged from 59% (colorectal) to 74% (lung). When the aggregate of cases found by either Medicare or Registry were used the capture rates ranged from 74% (prostate) to 89% (breast). Further analysis indicated that capture rates differed by demographic characteristics.

CONCLUSION

We conclude that Medicare claims are useful to supplement a Registry, estimate the number of missing cases, and assess bias in capture.

Pharmacy and medical claims data identified migraine sufferers with high specificity but modest sensitivity

OBJECTIVE

Claims data are often used to identify and monitor individuals with particular conditions, but many health conditions are not easily recognizable from claims data alone. Patient characteristics routinely available in claims data were used to develop model-based claims signatures to identify migraineurs.

STUDY DESIGN AND SETTING

A validated telephone interview was administered to 23,299 continuously enrolled managed care members aged 18-55 to identify 1,265 migraineurs and 1,178 controls. Responses were linked to medical and prescription claims. Claims variables were evaluated for sensitivity, specificity, and positive and negative predictive value in predicting migraine status. Regression models for predicting migraine status were developed.

RESULTS

Regression-based claims signature models were successful in case-finding, as indicated by fairly sizable odds ratios (OR). In the full model (including demographic, medical, pharmacy, and comorbidity claims variables), a claim for a migraine drug, gender, and a claims-based headache diagnosis were strongly associated with migraine case status (OR=3.9, 3.2, and 3.0, respectively).

CONCLUSION

Using either medical or pharmacy claims provided highly specific and moderately sensitive case-findings. Strategies that combined medical and pharmacy information improved sensitivity and may increase the usefulness of claims for identifying migraine and improving the quality of migraine care.

The sensitivity of Medicare billing claims data for monitoring mammography use by elderly women

Mammography use is monitored through Medicare billing claims; however, the sensitivity of this data source has not been previously described. This study included 10,852 Colorado women ages 65 and older with a mammogram in 1998 as registered by the Colorado Mammography Project who were Medicare fee-for-service (FFS) enrollees. These records were matched to Medicare billing data to assess the proportion of those mammograms submitted for payment to Medicare. The overall sensitivity of the FFS Medicare billing data for screening mammography was 85 percent. Medicare billing claims were less sensitive for younger women, African Americans, women with some college education, and women with supplementary private insurance. In Colorado, the Medicare FFS billing claims understates mammography usage by 15 percent. Care must be taken when comparing mammography use derived from Medicare billing claims, as the sensitivity of billing data can vary substantially by age, race, and socioeconomic status.

Evaluation of an algorithm to identify incident breast cancer cases using DRGs data

Hospital databases have the potential to be inexpensive, timely and nationally representative sources of information about cancer. This study examines the utility of the French hospital database adapted from the Diagnosis Related Group (DRG) classification and named 'Programme de médicalisation des systèmes d'information (PMSI)', as an independent source to identify incident cancer cases. From the 19 679 women hospitalized and treated in 1998 in the public hospitals of the Somme area in France, we identified those diagnosed with breast cancer in the PMSI database. These women were matched with women in the cancer registry of the Somme area who had been diagnosed with breast cancer in 1998. An algorithm was used to identify cancer-related diagnoses and procedures reported to PMSI. The sensitivity, specificity and positive predictive value (PPV) of the PMSI database were calculated using the cancer registry as a gold standard. The PMSI database had 85% sensitivity, 99.9% specificity and 97% PPV for women hospitalized with breast cancer as a principal diagnosis. The sensitivity was higher by 9% for hospitalization with breast cancer as a secondary diagnosis but had a lower PPV (78%). In conclusion, the PMSI database seems to offer an interesting potential to assess breast cancer incidence, because of its high sensitivity, in particular when secondary diagnosis was considered, and its very high specificity and PPV. However, these preliminary results need to be confirmed by other studies in France before such databases are used, particularly in areas without cancer registries.