Amish Study of Major Affective Disorders
Janice Egeland, Ph.D.

The Amish Study of major affective disorders (bipolar and unipolar forms) began in 1976 with grant support from the National Institute of Mental Health (Egeland and Hostetter, 1983). The study was facilitated by Dr. Janice A. Egeland, the principal investigator, who had been observing and studying the health of this population since 1960 and documenting large families with mental illness. Several methods were designed for the formal ascertainment of mentally ill subjects in multigenerational pedigrees and for clinical evaluations and diagnoses (see below). Pedigrees for genetic study were based on type I bipolar disorder (BPI) probands and on the clustering of psychopathology in the extended families. This process has been continuous for a 32-year period (1976-2008). Blood samples were collected first for biochemical and genetic marker purposes commencing in 1977. Blood samples were also submitted to the NIGMS Human Genetic Cell Repository at the Coriell Institute for Medical Research (CIMR) in order to develop an Amish Study Collection for bipolar affective disorders.

Unique Research Setting

Population isolates are well known for their genetic and cultural homogeneity and their value in medical genetic and molecular biologic study (McKusick et al., 1964). The Old Order Amish, the most conservative group, is characterized by their large families, clear paternity, extensive genealogic records, and geographic immobility. They were willing to support this medical study because of its potential benefits for all people with a "manic-depressive" disorder. The Old Order Amish in this research reside primarily in Lancaster County, Pennsylvania, and represent the oldest deme in the United States. It was founded in pre-Revolutionary days by a small number of Swiss immigrants. It has remained essentially a closed, endogamous group numbering some 12,500 when the research began and had doubled in size by 2000. Relatives of BPI probands have been followed to complete the CIMR cell collection in secondary settlements of Ohio, Indiana, New York, Maryland, Kentucky, and Florida. As an inbred population, this represents a rich resource for genetic studies of rare recessive disorders, dominant conditions, complex traits, chromosomal aberrations, the role of assortative mating, consanguinity, and gene-environment interactions. Over the decades, Amish pedigrees have been cited as excellent general reference cell collections for characterization of new genetic probes and the construction of the human genetic map.

Psychiatric Ascertainment and Diagnosis

Ascertainment was a process completely separate from clinical evaluation. Cases were reported through a community wide network of Amish scribes in every church district (representing the number of families worshipping together). Scribes were given guidelines to help with recognition of significant signs/symptoms of mental illness and also reported any hospitalization or treatment for a psychiatric problem. It was typical to learn of serious mental illness from multiple other informants, basically close family and church members.

Two methods have been used for collection of clinical information: psychiatric interviewing and abstraction of medical records. The Amish Study staff included persons trained in administering the Schedule for Affective Disorders and Schizophrenia – Lifetime Version (SADS-L). It was pre-tested to confirm that the vocabulary and context of questions was culturally appropriate. An interviewer read SADS-L questions as another took detailed notes. A pre-coded checklist was not used as a basis for diagnosis. The interpretation of interview answers was left to professional psychiatrists. Policy required interview of the patient, if possible, and at least three of the closest family members, with the average being six. The second method was abstraction of all medical records (hospital, clinic, private physician) using informed consent, both IRB-based and those from local psychiatric facilities. The issue of "blindness" was strategic and included: no names, no references to existing diagnoses, no relationship to other subjects, no specific medications/treatment, any of which could influence an independent diagnosis. These same procedural methods remain in place (Hostetter et al., 1983; Egeland et al., 1990).

Reliable diagnosis is crucial to a psychiatric genetic study. All patients were diagnosed by a five member Psychiatric Board using the Research Diagnostic Criteria (RDC) and the latest version of the Diagnostic and Statistical Manual of Mental Disorders (DSM III-IV) which give uniform guidelines for clinical assessment. One-third of all BPI patients were diagnosed blindly at two points in time, once with interview materials and another time using medical records. There was complete concordance. The stability of diagnoses over time (validity) was documented by the fact that most cases were re-evaluated by subsequent episodes. Measures of inter-rater reliability of diagnoses were reported for the clinicians. A final consensus diagnosis was a hallmark of the Amish Study because it did not depend on the idiosyncratic diagnosis of any one clinician and avoided the problem of drift in diagnosis over time. After confirmation of a BPI proband by the project Psychiatric Board, all first and many second degree relatives were interviewed using the Family History / Research Diagnostic Criteria (FH/RDC) method and by a full SADS-L interview, if warranted. The Amish Study has followed individuals longitudinally to note new onset of illness and changes in the course-of-illness. The psychiatric status of individuals represented in the NIGMS Human Genetic Cell Repository has been updated as of 2008 (Hostetter et al., 1983; Egeland et al., 1990).

Amish Study Genetic Linkage Research

A rich database has been assembled for the Amish pedigrees included in this catalog. Dr. Egeland and her colleagues have published data on selected biochemical variables and a number of conventional genetic markers typed in these pedigrees (Kidd et al., 1984; Egeland et al., 1984; Goldman et al., 1985). Typing of these individuals demonstrated that considerable heterozygosity existed for the classical markers (red cell antigens and serum proteins). Amish Study collaborators were also conducting chromosome loci linkage searches (Ginns et al., 1996; Ginns et al., 1998). Recently, single nucleotide polymorphisms (SNPs) and STR/microsatellites have been typed, with an extensive genomic scan and large-scale sequencing underway. Data will be available to other scientists in the future.

The genealogic connection between pedigrees has been traced. Cell lines have been established for numerous additional patients and non-affected relatives extending into a fourth and even a fifth generation. Children at "high risk" for bipolar disorder are being evaluated in a prospective study (CARE Program) conducted since 1994 (Egeland et al., 2003; Shaw et al., 2005). Each informative person added to such a kinship increases substantially the power of linkage detection. Extending this valuable national resource is highly desirable for both psychiatric linkage purposes and as general reference material.

Pedigree Selection: 110/210/310/410

Pedigree 110 (NIGMS Repository Family 884) was first identified as a key psychiatric resource at an NIMH workshop of genetic consultants to the Amish Study in January 1978. It was cited because of clustered psychopathology in large kinship lines with three generations available for blood marker research: parents, children and grandchildren. The sibships numbered eight or more showing a pattern of illness. The unique value of Pedigree 110 as a resource for other investigators worldwide led to the first development of permanent cell lines in 1983 at the NIGMS Repository. Lymphoblastoid and fibroblast cell lines were established initially from 51 individuals. Blood samples were drawn after full disclosure of the purpose of the study and with two signed consent forms, approved by the IRB of the University of Miami.

Additional lymphoblastoid cell lines were established in following years. Highest priority was given to those with confirmed bipolar affective disorder and their "at-risk" children and siblings. This original affective disorder cell collection numbered 81. A companion pedigree, Pedigree 210 (NIGMS Repository Family 1075) linked to Pedigree 110 (Family 884) by marriage and consanguineous ties, was introduced to the collection by tissue samples. As of 2008, the subjects that have cell lines currently available for Pedigrees 110 and 210 total 235.

Two additional large bipolar pedigrees have been developed over the past 15-year period of ascertainment and diagnostic evaluation by the Amish Study network. They are Pedigree 310 and Pedigree 410, providing an additional 246 individuals. Transfer of these materials to the NIGMS Repository was completed in 2007, with the latest diagnostic status for all 481 subjects as of 2008.


Egeland JA, & Hostetter AM: Amish Study: I. Affective disorders among the Amish, 1976-1980. Am. J. of Psychiatry 140:56-61, 1983.

Egeland JA, Sussex JN, Endicott J, Hostetter AM, Offord DR, Schwab JJ, Allen CR & Pauls DL: The impact of diagnoses on genetic linkage study for bipolar affective disorders among the Amish. Psychiat Genet, 1:5-18, 1990.

Egeland JA, Kidd KK, Frazer A, Kidd J, & Neuhauser VI: Amish Study: V. Li-Na counterflow and COMT in bipolar pedigrees. Am. J. of Psychiatry 141:1049-1054, 1984.

Egeland JA, Shaw JA, Endicott J, Pauls DL, Allen CR, Hostetter AM & Sussex JN: Prospective study of prodromal features for bipolarity in well Amish children. J. Am. Acad. Child Adolesc. Psychiatry, 42:7, 786-796, 2003.

Ginns E, Ott J, Egeland J, Allen C, Fann C, Pauls D, Weissenbach J, Carulli J, Falls K, Keith T & Paul S: Genome-wide search for chromosomal loci linked to bipolar affective disorder in the Old Order Amish. Nature Genet, 12:431-435, 1996.

Ginns E, St. Jean P, Philibert R, Galdzicka M, Damschroder-Williams P, Thiel B, Long R, Ingraham L, Dalwaldi H, Murray M, Ehlert M, Paul S, Remortel B, Patel A, Anderson M, Shaio C, Lau E, Dymarskaia I, Martin B, Stubblefield B, Falls K, Carulli J, Keith T, Fann C, Lacy L, Allen C, Hostetter A, Elston R, Schork N, Egeland J & Paul S: A genome-wide search for chromosomal loci linked to mental health wellness in relatives at high risk for bipolar affective disorder among the Old Order Amish. Proc. Natl. Acad. Sci. USA, 95:15531-15536, 1998.

Goldman D, Goldin LR, Giri PR, O’Brien SJ, Egeland JA, & Merril CR: Twenty-seven protein polymorphisms by two-dimensional electrophoresis of serum, erythrocytes, and fibroblasts in two pedigrees. Am. J. of Human Genet. 37:898-911, 1985.

Hostetter AM, Egeland JA, & Endicott J: Amish Study: II. Consensus diagnosis and reliability results. Am. J. of Psychiatry 140:62-66, 1983.

Kidd KK, Egeland JA, Molthan L, Pauls DL, Kruger SD, & Messner KH: Amish Study: IV. Genetic linkage study of pedigrees of bipolar probands. Am. J. of Psychiatry 141:1042-1048, 1984.

McKusick VA, Hostetler JA, & Egeland JA: Genetic studies of the Amish: Background and potentialities. Bulletin of Johns Hopkins Hospital 115(4):203-222, 1964.

Shaw JA, Egeland JA, Endicott J, Allen CR, Hostetter AM. A 10-year prospective study of prodromal patterns for bipolar disorder among Amish youth. J. Am. Acad. Child Adolesc. Psychiatry, 44:11, 1104-1117, 2005.

Amish Study of Major Affective Disorders main page