MicSem Publications

WHO Mental Health Research in Micronesia

by Francis X. Hezel, S.J.

2003 Health

This review of mental health research covers those parts of the former Trust Territory of the Pacific Islands that have since 1986 become independent nations: namely, the Federated States of Micronesia (including the four states of Yap, Chuuk, Pohnpei and Kosrae), the Republic of Palau, and the Republic of the Marshall Islands.

Mental Health services in Micronesia, which were first organized under the Trust Territory in 1974, have a history of less than 30 years, while research extends back only 20 years. Inasmuch as the history is so short, this report will encapsulate all the attempts at mental health research with which I am familiar, however limited the scope of the work.

Please note that the dates in parentheses next to the name of the author indicate the time frame for the data used rather than publication date of the work.

1. HBM Murphy (1977-1978).

Murphy records admission rates for mental illness during the early 1970s as ranging from 0.1 per thousand adults in the Solomon Islands, to 0.2 or 0.3 in Tonga and Western Samoa respectively, to 0.8 in the Gilbert Islands, to 2.2 among Melanesians in New Caledonia. In comparing these admission rates with those of industrialized nations in the West, Murphy concludes that the admission rate for the latter "was running at least five times higher than the average for territories" (Murphy, 1978, p. 5).

Acknowledging this difference may reflect the fact that fewer Pacific Islanders have access to mental health facilities than Westerners, Murphy attempts to correct this bias by comparing admission rates for Islanders living close to hospitals and those living in more remote areas. On the basis of comparative data from four places (Fiji, New Hebrides, the Solomons, and Tonga) showing that hospitalization rates are actually higher in some of the most remote areas, Murphy discounts this bias. He does not, however, dispose of another, perhaps even more important bias-the well-known tendency of Pacific Islanders to care for the mentally ill at home rather than refer them to the hospital. Hence, while dealing with the problem of access to hospital care in the Islands, Murphy recognizes but does not respond to the problem of cultural preference in the care of the mentally ill.

Through field surveys conducted in several small communities by key informants, Murphy estimates the rate of psychosis at 2.5 per thousand adults. (He estimates the schizophrenia rate at 2.0 per thousand.) When compared with the average of 9 per thousand reported for the industrialized world, Murphy concludes that the prevalence rate of serious mental illness in the Pacific is about one-fourth of what it is in the West. Murphy attributes the significant difference in rates to the higher levels of stress concomitant with "modernization." As he puts it, "rates (of mental illness) appear lowest where the percentage of the population following a subsistence economy is highest, and tend to increase as the percentage in the cash economy rises" (Murphy, 1978, p. 37). He adduces as supporting evidence the relatively high rates of illness in Guam and New Caledonia, two of the most modernized places he visited. For Murphy the root explanation for the low rates in the Pacific lies in the communal lifestyle found throughout the Pacific. This communal way of life affords the individual ample emotional support, frees him from the need to make difficult decisions on his own, buffers the symptoms of those with mental disorders, and diminishes the need for a strong superego, thus reducing the risk of ego-superego conflicts.

Murphy notes what will be a continuing theme in epidemiological literature on the Pacific: the preponderance of males over females found to be suffering from mental illness. Among local populations, the male rate of those admitted for hospitalization is twice as high as the female rate. This he explains by appealing to cultural differences in symptomatology, with males normally behaving much more aggressively than females in accordance with cultural norms.

Whatever we may think of Murphy's theoretical framework, he stands as one of the pioneers in the epidemiology of mental illness in the Pacific. Furthermore, his work is wide-ranging enough to include such important phenomena as spirit possession states, epidemic hysteria, and suicide. Although his epidemiological data, by his own admission, is weak, he paints a large-canvas picture of the psychological problems faced by Islanders then and since.

2. Epidemiological Snapshots (1976-1983)

During the declining years of the trusteeship in Micronesia, a handful of short articles drawing on figures from the authors' case registries appeared.

LG Wilson, in a brief article on his experience as Trust Territory psychiatrist, offers a summary of psychiatric evaluations of 114 Micronesians, presumably done by himself, during 1976-1977 (Wilson, 1980). Of the 114 persons seen, 79 were diagnosed as psychotic: 57 with schizophrenia, 6 with bipolar disorder, 9 with psychotic depression, and 7 with other psychoses. The remainder suffered from organic mental disorders, mental retardation, drug use, or neuroses. This information is of limited epidemiological value since the author offers no breakdown by gender or island group.

Andrew Allan and Ernest Hunter, both teaching in the Department of Psychiatry at the University of Hawaii School of Medicine, assisted in a six-month mental health training program for Micronesian medical officers in 1978. Following the completion of this program, the authors were engaged in follow up visits to Micronesia during a three-year period 1980-1983. The data in their paper were drawn from the psychiatric consultations the pair conducted during this three year period with 37 individuals referred to them (Allan & Hunter, 1985). According to their diagnoses, only 13 of these 37 persons suffered from psychosis-in all cases schizophrenia. All but one of these diagnosed schizophrenics were male. Six of the 13 were from the Marshall Islands, an island group that Paul Dale, who was doing his own epidemiological study at about the same time, claimed to have a very low rate of schizophrenia (Dale, 1981). Allan and Hunter, who also examine the reason for referral, conclude that most of the males (74%) presented because of social disruption, while most of the females (64%) were referred due to subjective distress.

3. Paul Dale (1978-1979)

Paul Dale, who served as regional psychiatrist for Micronesia, also drew on his case registry for 1978-1979 to provide the data used in his 1981 article, "Prevalence of Schizophrenia in the Pacific Island Populations of Micronesia" (Dale, 1981). He used data from the monthly reports of mental health services to support his thesis that, contrary to general belief in the psychiatric field, the rates of schizophrenia varied greatly from one part of Micronesia to another.

Prevalence of Schizophrenia and Chronic Mental Illness (Dale 1981)

Island Group Schizophrenia # Cases Other Psychoses # Cases Schizophrenia Specific Rate All Psychoses Specific Rate
Palau 63 15 7.6 9.4
Yap 40 10 6.7 8.3
Chuuk 40 4 2.2 2.4
Pohnpei 17 8 1.4 1.9
Kosrae 3 1 1.2 1.6
Marshalls 11 1 0.8 0.9
Total 174 39 2.8 3.4

Dale notes the wide range of prevalence rates for schizophrenia from one island group to another, with Palau having a rate about ten times as high as the Marshalls. The rates for islands, listed from west to east on the table above, show a declining rate as one moves eastward. Dale further notes that the island populations of two Polynesian outliers, Nukuoro and Kapingamarangi, both part of Pohnpei State, had no known cases of schizophrenia, although there was one person (of a combined adult population of 1,044) suffering from bipolar psychosis. Dale uses this as additional evidence to support his conclusion that schizophrenia, with rates varying greatly from place to place, "is not a universal affliction on mankind" (Dale, 1981, p. 111].

In a methodological note, Dale states that "all persons were examined by the same professionals using the same criteria in all islands" (Dale, 1981, p. 105). Although making use of the records of Mental Health Services, he seems to suggest that he or another professional psychiatrist interviewed all those diagnosed as schizophrenics. He claims that reliance on mental health records was supplemented by a survey conducted in each place to seek out the mentally ill, but it is unclear just how this was done.

4) Hammond, Kauders and MacMurray (1981)

One of the authors, Frank Kauders, a psychiatric resident from Loma Linda Medical School, conducted a three-month study of schizophrenics in Palau during his three-month rotation there (Hammond, Kauders & MacMurray, 1983). Using Mental Health Services case files, he identified 73 schizophrenics in Palau. This would yield a rate of 8.8 per thousand adults (rather than the rate of 4.6 that he cites in his article). He notes the greatly skewed gender distribution; 58 of the patients were males and 15 were females for a 4:1 ratio.

During his three-month stay in Palau, Kauders interviewed 35 of those persons identified as schizophrenics. In studying the symptoms of the disease, he found a proclivity toward violence among the predominantly male patients. He and his colleagues found a high correlation with rampant drug use, especially cannabis and alcohol. While also noting common themes in the life histories of patients-60% had spent time away from Palau, for instance-they attributed the high male rate of schizophrenia to widespread drug use among males and to the rapid disintegration of the traditional culture. Their assumption was that such changes in the social environment would affect men more deeply than women, although they do not explain why. They seem to further suggest that the use of cannabis by males has intensified the symptoms of the disease, sometimes even misleading observers into a misdiagnosis of what is in reality a temporary condition that the authors call "cannabis psychosis."

5) Hezel and Wylie (1989-1990)

Francis Hezel, a long-time resident and social researcher in Micronesia, attempted his own survey of serious mental illness in 1988. Working with educated lay field workers and using Mental Health Office case files, he and his team conducted a survey over an 18-month period. They interviewed community leaders and other key informants to identify all persons, whether they had ever sought treatment or not, who had acted "crazy" for a year or longer. The criteria for psychosis, then, were based on emic community norms. Excluded on the basis of further questioning were organic disorders, congenital mental retardation, epilepsy, personality disorders, possession syndromes, and temporary conditions. With the collaboration of Michael Wylie, then a psychology professor at the University of Guam, Hezel published the results of the Micronesian-wide survey (Hezel & Wylie, 1992).

Because community norms for mental illness rather than strict psychiatric evaluations are used, the findings of this study do not present a fine-tuned view of the prevalence rate of schizophrenia as such. It can, however, afford us a reasonably accurate picture of the prevalence of chronic psychosis in the area and of the comparative rates in the various island groups.

The survey revealed a total of 445 individuals suffering from psychosis and yielded an overall rate of chronic mental illness for Micronesia of 5.4 per thousand. This represents a 58% increase over the rate recorded by Dale ten years earlier. It is unclear, however, whether this represents an actual increase in the prevalence rate, or whether the higher numbers are due to better case-finding methods or a broader definition of chronic mental illness. The rates obtained in this study fall within the accepted prevalence ranges established across cultures (eg, in the WHO Ten Country Study of 1992). Although the overall rate of 5.4 is higher than Murphy's estimated psychosis prevalence rate of 2.5 in the late 1970s, it seems to be on the low end of the rate linked with most Western nations.

The survey also confirmed the considerable differences in prevalence rates from one island group to another. Palau once again showed the highest rate of psychosis at 16.7; its rate was almost double that of the next highest island group, Yap (8.4). The prevalence rates in the other groups ranged between 6.4 (in Kosrae) and 3.2 (in Pohnpei).

Prevalence of Schizophrenia and Chronic Mental Illness (Hezel & Wylie 1992)

Island Group Male Rate Female Rate Total Rate
Palau 22.5 10.5 16.7
Yap 12.1 4.6 8.4
Chuuk 6.5 1.1 3.8
Pohnpei 4.8 1.6 3.2
Kosrae 12.8 0 6.4
Marshalls 6.5 1.9 4.2
Total 8.4 2.5 5.4

This survey confirmed the striking difference between male and female rates that had been noted in previous work. The overall male/female ratio was 3.4/1, a ratio that was roughly reflected in the gender breakdown of the psychotic population of each island group. Consequently, specific rates for some of the western island groups were very high, with the rate for Palau males reaching 22.5. The prevalence rates seem to decrease with age for males, while they increase for females.

The life history data collected in the course of this survey allowed the authors to examine some of the characteristics of the psychotic population. The surveyed population is better educated than the general population, and over half of them have lived abroad for six months or longer. A large percentage (88%) of all psychotic males have had a history of significant drug and/or alcohol use, while only 36% of the females have a similar history. The authors conclude that preponderance of male morbidity is real rather than an artifact of exaggerated symptoms in males by comparison with females. They argue that it is attributable to environmental factors, including higher exposure of males to drugs and to role changes with their concomitant stress.

6) Marina Myles-Worsley et al (1998)

Myles-Worsley and her colleagues from the University of Utah School of Medicine chose Palau, with its limited population and its high rate of schizophrenia, to study the genetic etiology of schizophrenia (Myles-Worsley et al, 1998). This study signals a marked shift in research interest away from broader epidemiological concerns, including the socio-cultural factors responsible for the disease, to genetic transmission of the illness. This was only to be expected in view of the race to map the gene and identify those genomes responsible for different human traits.

The first phase of the project and the one of most interest to us was to establish the lifetime prevalence of schizophrenia in Palau. With the help of the staff of Mental Health Services in Palau and the use of case files and community informants, the researchers were able to identify 160 strictly defined cases of schizophrenia in the island group. This yielded a lifetime prevalence rate of 20 per thousand overall, 28 for males and 12 for females. The authors point out that such elevated rates of schizophrenia are to be found in a few isolate populations, such as Northern Sweden and a section of Finland, where the lifetime prevalence rates range between 20 and 30 per thousand.

The study, like many previous works, reports on the difference between male and female rates in Palau, with males running twice the risk of schizophrenia. Moreover, male onset occurs an average of 4 years earlier than female onset, as other studies have noted as well.

After reviewing the familial links between those with schizophrenia, the authors conclude that a likely explanation of the high prevalence rate in Palau is the aggregation of "minor susceptibility genes of low penetrance because of the large sibships and half-sibships produced by Palauan marriages" (Myles-Worsley et al, 1998, p. 5).


Allan, A. T., & Hunter, E. M. (1985). Cross-cultural psychiatry in Micronesia: The consultant's view. International Journal of Social Psychiatry, 31, 59-66.

Dale, P. W. (1981). Prevalence of schizophrenia in the Pacific Island populations of Micronesia. Journal of Psychiatric Research, 16, 103-111.

Hammond, K. W., Kauders, F. R., & MacMurray, J. P. (1983). Schizophrenia in Palau: A descriptive study. International Journal of Social Psychiatry, 24, 161-170.

Hezel, F. X., & Wylie, M. A. (1992). Schizophrenia and chronic mental Illness in Micronesia: An epidemiological survey. ISLA: A Journal of Micronesian Studies, 1:2, 329-354.

Murphy, H. B. M. (1978). Mental health trends in the Pacific Islands. Report on a tour of Pacific territories, September 1977-March 1978. South Pacific Commission, Noumea, New Caledonia.

Myles-Worsley, M., Coon, H., Tiobech, J., Collier, J., Dale, P., Wender, P., Reimherr, F., Polloi, A., & Byerley, W. (1998). A Genetic epidemiological study of schizophrenia in Palau, Micronesia: Prevalence and familiality. American Journal of Medical Genetics, 81, 1-7.

Wilson, L. G. (1980). Community psychiatry in Oceania: Fifteen months' experience in Micronesia. Social Psychiatry, 15, 175-179.