Original reports
A Retrospective Cohort Study of Implanted Medical Devices and Selected Chronic Diseases in Medicare Claims Data

https://doi.org/10.1016/S1047-2797(00)00037-5Get rights and content

Abstract

PURPOSE: Several case-control studies have observed associations of implanted medical devices and certain connective-tissue and neurologic diseases. We reexamined these and other associations using cohort comparisons.

METHODS: We compared the incidence of 52 diseases in several retrospective cohorts constructed from Medicare claims data. Six cohorts were defined by implantation of medical devices (silicone, metal bone or joint implants, breast implants, penile implants, pacemakers, artificial heart valves), and four comparison cohorts were defined by surgeries not involving implants.

RESULTS: We observed associations that were generally consistent with previous reports, including associations of bone and joint implants with connective-tissue diseases, and an association of penile implants with idiopathic progressive neuropathy. We also observed associations of breast implants and pacemakers with connective-tissue diseases.

CONCLUSIONS: For the most part, our study confirms our previous case-control results. Although confounding by presurgical conditions (such as diabetes) remains a plausible explanation of the findings, several associations are worthy of more detailed research.

Introduction

The continuing controversy surrounding effects of silicone breast implants 1, 2, 3, 4, 5, 6, 7 has motivated several studies of other implanted medical devices 6, 7, 8, 9. A University of Michigan group conducted population-based case-control studies of the association of implanted devices with scleroderma and with undifferentiated connective-tissue disease (UCTD), defined as International Classification of Diseases—Version 9 (ICD-9) 710.9. Due to the small numbers of exposed cases, results from the scleroderma study were ambiguous (7), but associations of internal fixation devices, artificial joints, and pacemakers with UCTD were detected (6).

In case-control studies of private and Medicare claims databases, we detected associations of bone, joint, and penile implants with certain connective-tissue and neurologic diseases 8, 9. In the second study, we also examined several non-implant surgical procedures as “surgery-severity” controls for implant procedures: carpal-tunnel surgery for comparison to small-joint implants, hernia repair for comparison to large-joint implants, and trans-urethral prostatectomy (TURP) for comparison to penile implants (9). Carpal-tunnel surgery exhibited associations with connective-tissue and neurologic diseases at least as strong as those seen for implants, suggesting that pre-existing deterioration might explain the implant associations. Nonetheless, unlike with large-joint and penile implants, we detected no positive associations that involved hernia repair or TURP (9).

In light of the above findings, we decided to modify and expand our previous analysis of the Medicare claims database (9). The modification involved adopting a retrospective cohort rather than case-control comparison. Because the eligible Medicare population is not enumerated in the claims data, it is not possible to define population-based controls (noncases) or untreated cohorts from these data without using Medicare “denominator” files. Therefore, in our case-control studies we chose controls from a spectrum of diseases that appeared to be unassociated with implants 8, 9. To address concerns that these control choices might be partly responsible for our earlier results, we constructed and compared cohorts defined by medical procedures.

Compared with our Medicare case-control study (9), four expansions were made here. First, we examined more implanted devices, including pacemakers (which were associated with UCTD in an earlier study (6)). Second, we examined more control procedures, including arthroscopic surgery (which shared some indications with joint-implant surgery). Third, we examined additional symptoms and diseases that have possible autoimmune or neurologic involvement. Fourth, we included two more years of data, which provided needed power and precision for the cohort analysis.

Section snippets

Population and Database

The Medicare population comprises US citizens covered by the Medicare program. The Health Care Financing Administration (HCFA) provides a 5% sample of the claims files of the insured. The data in the present study were obtained from the HCFA samples for 1991 to 1996 (inpatient, outpatient, and physician/supplier part B). The files contain age, sex, outpatient and inpatient procedures, and dates of occurrence, so that patients with implants may be identified by dates of procedures. Diagnoses for

Presentation Screening

Because our analyses produced 312 rate-ratio estimates with accompanying confidence limits and p-values (six nonreference procedures × 52 diagnoses) for a total of 312 · 4 = 1248 statistics, we had to set explicit rules to reduce the number of statistics for presentation in this article. Any presentation rule is bound to be somewhat arbitrary. We, therefore, chose to use a liberal inclusion criterion, so that most readers will be able to find estimates of special interest. All our analysis

Comparisons with Past Findings

As mentioned earlier, special care is needed in interpreting our results in relation to earlier findings, primarily because of our use of surgical comparison cohorts. Our comparisons were motivated by the observation that carpal-tunnel surgery exhibited strong associations with connective-tissue and neurologic conditions, much like those seen for silicone bone and joint implants (9). We chose arthroscopic surgery for our new comparisons for several reasons: it is very common, and so would yield

Conclusions

Using a different sampling design and new comparison groups, we have again observed the associations of implanted devices with connective-tissue diseases seen in our case-control studies 8, 9. Like previous device studies 6, 7, 8, 9, ours may have been confounded by pre-existing medical conditions. Because the diseases in question are rare, a very large population source (with hundreds of thousands of person-years of exposure) is needed to construct an informative study. At the same time, to

Acknowledgements

We are grateful to Peter Alexakis, M.D., Chris Hafeman, M.B.A., R.R.A., Molly Neal, C.C.S., Edward O'Connor, M.D., and Andrew Spitzer, M.D. for consultations regarding surgeries and implants. This work was supported by Dow-Corning Corporation.

References (25)

  • S. Greenland et al.

    A case-control study of prosthetic implants and selected chronic diseases

    Ann Epidemiol.

    (1996)
  • S. Greenland et al.

    A case-control study of prosthetic implants and selected chronic diseases in Medicare claims data

    Ann Epidemiol.

    (1998)
  • B.G. Silverman et al.

    Reported complications of silicone gel breast implantsAn epidemiologic review

    Ann Int Med.

    (1996)
  • F. Wolfe

    Silicone breast implants and the risk of fibromyalgia and rheumatoid arthritis

    Arth Rheum.

    (1995)
  • C.H. Hennekens et al.

    Self-reported breast implants and connective-tissue diseases in female health professionals

    JAMA.

    (1996)
  • D. Friis et al.

    Connective tissue disease and other rheumatic conditions following breast implants in Denmark

    Ann Plast Surg.

    (1997)
  • H.J. Williams et al.

    Breast implants in patients with differentiated and undifferentiated connective-tissue disease

    Arth Rheum.

    (1997)
  • T.J. Laing et al.

    The association between silicone exposure and undifferentiated connective tissue disease among women in Michigan and Ohio

    Arth Rheum.

    (1996)
  • C.J. Burns et al.

    The epidemiology of scleroderma among womenAssessment of risk from exposure to silicone and silica

    J Rheumatol.

    (1996)
  • J.L. Kelsey et al.

    Methods in Observational Epidemiology

    (1996)
  • S.R. Searle et al.

    Variance Components

    (1992)
  • N.E. Breslow et al.

    Approximate inference in generalized linear mixed models

    J Am Stat Assoc.

    (1993)
  • Cited by (32)

    • Appropriateness of Antibiotic Prophylaxis Before Dental Procedures, 2016–2018

      2022, American Journal of Preventive Medicine
      Citation Excerpt :

      Appropriate antibiotic prophylaxis was defined as a prescription dispensed before a dental visit with a procedure that manipulated the gingiva or tooth periapex or perforated the oral mucosa in patients with an appropriate cardiac diagnosis as indicated by the guidelines.3 Patients were identified as having a prosthetic joint if they had a hospital and medical claim of a prosthetic joint implant or complication before the antibiotic dispense date.6–8 Descriptive characteristics associated with dental visits were age at visit, sex, region in U.S., rurality, antibiotic associated with the visit, category of dental procedure, previsit conditions (history of prosthetic joint, cardiac condition, diabetes, immunocompromised, etc.), preindex health service use (primary care provider visits, specialist visits, emergency department visits, and hospital admissions).

    • 7.9 Orthopedic implant use and infection

      2017, Comprehensive Biomaterials II
    • Routine health insurance data for scientific research: Potential and limitations of the Agis Health Database

      2011, Journal of Clinical Epidemiology
      Citation Excerpt :

      The potential of such studies was illustrated even before the automation era by a study of the mortality of workers with asbestos exposure in 1935 [4]. More recently, they have proven to offer excellent opportunities to study effects of interventions and, in particular, side effects of drugs [5,6]. Health Insurance Databases (HIDs) such as those of the Health Insurances Companies in the Netherlands, Krankenkasse in Germany, and Medicaid in the United States have a high reliability because of their background of financial reimbursement of health care deliverables to care providers.

    • A review of uses of health care utilization databases for epidemiologic research on therapeutics

      2005, Journal of Clinical Epidemiology
      Citation Excerpt :

      Such limitations make it inevitable that epidemiologic research is performed post marketing to define these issues [1]. Although the focus of pharmacoepidemiology is on post-marketing surveillance of drugs, biologics [2], and medical devices [3], the approach has valuable applications in the pre-marketing phase to assess the safety profile of drugs and put them into context of the natural history of the condition they are designed to treat [4]. Although pharmacoepidemiology makes use of all epidemiologic study designs and data sources, in recent years there has been enormous growth in the use of large health care databases [5].

    • Mortality among semiconductor and storage device-manufacturing workers

      2005, Journal of Occupational and Environmental Medicine
    View all citing articles on Scopus
    View full text