Now that we recognize the role and the need for information about the natural course of illness and treatment, where can we find the information that we need? A reasonable place to begin any scientific inquiry is the literature and it is important to examine the standard clinical, epidemiological and social science literatures. Although not prominent in the mind of many pharmacovigilance professionals, we are often given hints from disparate sources that suggest additional details that might be relevant to the issue under study. The literature may also help guide the researcher to other articles, data sources or experts for further work. Beware however, that although "comorbidity" is a MeSH term, there is no equivalent for "disease natural history''.
Population-based studies are perhaps the most difficult but useful of all samples. These studies are undertaken in a population usually defined by geographic boundaries and has reasonable certainty for capturing any critical data of interest for that population. Perhaps the most well-known population-based resource is the Rochester Epidemiology Project, housed at the Mayo Clinic (Melton, 1996). The Rochester Epidemiology Project has a unique medical records linkage system that provides comprehensive capture of all care delivered to residents of Rochester and Olmsted County, Minnesota. Any medical care delivered to a resident of the county is reflected in their unique medical dossier and is available for research, which has produced over 900 publications since its organization in 1966. Olmsted County is one of the few places in the world where the occurrence and natural history of diseases can be accurately described and analyzed in a defined population for a half century or more.
Longitudinal study data are the key source of information to any researcher, particularly in pharmacovigilance. Virtually impossible to obtain, they do exist, mostly as longitudinal records of the subjects' interactions with the medical care system as in medical claims data or clinical databases. Some are population-based (i.e. the Rochester Epidemiology Project, Scandinavian registries), others represent stable clinical populations (i.e. the General Practice Research Database (GPRD)), while others are only as longitudinal as the subject's fiscal coverage of a particular reimbursement (i.e. a particular Health Maintenance Organization, Medicaid). Many longitudinal studies have been undertaken with a principal disease target (i.e. Atherosclerosis Risk in Communities, MONICA) or a general group under study (e.g. the Nurse's Health Study); however, in the course of the conduct of the study, other disorders are captured. Framingham stands as an example of a population-based longitudinal cohort that regularly screens their subjects for signs, symptoms and social factors to help us assemble a picture of disease natural history although it does not capture medical care interactions throughout the follow-up period. The popularity of the disease registry as a method for collecting natural history data is a testament to the power and utility of these data.
Such studies provide a rich resource of information in this field either through extant publications or through additional exploration of the data. The US government, as well as other governments, routinely undertake population studies and make the data available to the public for further analyses. One need only visit any country's center for health or population statistics website to see the extent of the data available. One of the more interesting and valuable of these is the National Health and Nutrition Examination Survey (NHANES) which every decade in the United States provides us with rich data on disease prevalence (incidence if you use the follow-up data), comorbidity, test results which allow the researcher to identify clinically detected and undetected cases of disease because of its capture of signs and symptoms as well as diagnoses (Mannino et al., 2000). Data on hospitalizations, outpatient visits, surgical procedures and more are often available. The Food and Drug Administration uses these data to generate the "expected rates'' of illness (La Grenade et al., 2000). Even long-term simplified trials that have become the mainstay of cardiovascular disease, can offer insight into the natural history of disease.
With the computerization of medical care, enormous aggregates of patient-level data have become available. Whether based on clinical encounters with a general practitioner (i.e. GPRD, MediPlus) or the claims from the encounter used in billing (most managed care databases, Medicare, Medicaid), much about practitioner recognized and treated disease can be understood. Analysis is also facilitated by the computerized nature of the data. The reader is referred to Chapters 30-32 in this same volume for more discussion about these resources.
Was this article helpful?
This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.