Prevalence of Psychiatric Disorders Among National Health Insurance Enrollees in Taiwan

1
Prevalence of Psychiatric Disorders Among
National Health Insurance Enrollees in Taiwan

• I-Chia Chien
• M.D., M.H.A., Ph.D.
• Tsao-Tun Psychiatric Center Institute of Public
  Health, National Yang-Ming University

2
Outline of presentation

• Background
• Literature review
• The aim of this study
• Material Methods
• Results
• Discussion
• Conclusions

3
Background

• National Health Insurance (NHI) program
  implemented in March 1995
• 96 of all residents of Taiwan in 2000 enrolled
  in the NHI program
• NHRI release the data for study in 2002
• Mental health will be worthwhile and emphasized
  in this centry
• Evidence-based study is emphasized

4
Literature reviewNational Health Insurance in
Taiwan

• Before the National Health Insurance (NHI) was
  introduced in 1995, 57 percent of Taiwans people
  were insured through three separate major social
  health insurance program, including Labor
  Insurance, Government Employee Insurance, and
  Farmers Insurance.

5
National Health Insurance in Taiwan

• Taiwan implemented a NHI program in March 1995,
  offering a comprehensive, unified, and universal
  health insurance program to all citizens.
• The coverage provided outpatient service,
  inpatient care, Chinese medicine, dental care,
  childbirth, physical therapy, preventive health
  care, home care, and rehabilitation for chronic
  mental illness.

6
National Health Insurance in Taiwan

• All citizens who have established a registered
  domicile for at least 4 months in the Taiwan area
  should be enrolled in NHI.
• Those individuals who do not have Taiwan
  citizenship but do have a Taiwan Alien Residence
  Certificate (ARC) must also participate,
  four-months after the ARC is issued, in the NHI
  program as of July 17th 1999.

7
National Health Insurance in Taiwan

• The insured are divided into six categories on
  the basis of their employment status and the
  group insurance applicant to which they belong.
• The premiums of the insured under Categories 4, 5
  and 6 are determined on the basis of the average
  premium of the whole group of the insured.

8
National Health Insurance in Taiwan

• The premiums of all the other insured are
  determined on the basis of the insured wage.
• The insured amount of the insured falling under
  Category 1 and Category is divided into 38 grades
  ranging from 15, 840 NTD to 87,600 NTD.

9
Catastrophic illness registration (CIR)

• ICD-9-CM codes
• 290 (Senile and presenile organic psychotic
  conditions)
• 293 (Transient organic psychotic conditions)
• 294 (Other organic psychotic conditions)
• 295 (Schizophrenic disorders)
• 296 (Affective psychoses)
• 297 (Paranoid states)

10
National Health Insurance in Taiwan

• As many as 96 of all residents of Taiwan
  (21,400,826) individuals, were enrolled in this
  health insurance system as of the year 2000.
• The Bureau of NHI (BNHI) contracted with 91 of
  the medical institutions in Taiwan.

11
Health care expenditures of Taiwan

• The total health care expenditures of Taiwan in
  2000 was US16.8 billion, equal to 5.36 of the
  gross national product (GNP)
• NHI expenditures in 2000 were US9 billion equal
  to 3 of Taiwans gross national product (GNP)
  that year.
• Expenditures on all psychiatric disorders
  combined represent only 3 of the NHIs budget
• (Department of Health, 2002).

12
Table1. Prevalence of psychiatric disorders in
previous studies in Taiwan
Includes mental retardation TPEP
Taiwan Psychiaric Epidemiologic Project MPM
Minor Psychiatic Morbidity NHI National Health
Insurance
13
Medicaid and Medicare study

• a sample of 5 of the Tennesseans (U.S.) with
  Medicare coverage
• only beneficiaries age 65 years and over were
  included in this analysis.
• 15.85 prevalence rate for all psychiatric
  disorders for the 3-year period from 1991 to 1993
  
• The findings suggested that race and gender
  interacted to influence service utilization and
  preventive care.
• (Husaini, 2000 Husaini, 2002).

14
Case finding strategy of psychiatric disorder

• Key informant method
• Field surveys
• Case registers

15
Effect of psychiatric disorder

• the presence of a diagnosable mental illness
  reduced employment by 11 for both males and
  females.
• For those who worked, the estimated loss of
  income attributable to mental illnesses was about
  20 for women and 10 for men (Ettner, 1997).

16
Economic costs of psychiatric disorder

• In Taiwan in 1994, the average annual economic
  costs of each patient with severe mental illness
  was NTD633,569, which amounted to approximately
  NTD50,000 each month.
• The annual costs for patients with schizophrenia
  and affective disorder was NTD680,889, and
  NTD412,738, respectively.
• (Yeh, 1997)

17
Economic costs of psychiatric disorder

• Among the total costs, 18.8 was direct costs
  while 78.3 was indirect costs. The indirect
  costs was higher than that of direct costs.
• They estimated that patients with schizophrenia
  spent 18 billion NTD and patients with affective
  disorder spent 45 billion NTD annually.
• (Yeh, 1997)

18
Treated rate of psychiatric disorder

• ECA study in United States revealed 29.5 people
  with psychiatric disorder.
• Only 20.9 among these psychiatric patients
  search for professional treatment within one
  year.
• General health care system offer service for 6.4
  people with psychiatric disorder
• Mental health care system offer service for 5.9
  people with psychiatric disorder.
• Most people with psychiatric disorder do not
  receive mental health service. (Reiger, 1993)

19
Accuracy of claim data

• If claim data are used for health care study, two
  sources of errors must be eliminated.
• The first is to establish the reliability of
  diagnosis between the claim data and patients
  chart.
• The second is to establish the validity of
  diagnosis that the patient does receive the right
  diagnosis (Lurie, 1992).

20
Accuracy of claim data

• Schwartz examined the reliability of psychiatric
  diagnoses among Medicaid enrollees. Reliability
  varies by diagnosis.
• The diagnosis of claim data agree with that of
  the chart in 50 to 94 of these cases (Schwartz,
  1980).
• Reliability is highest for diagnoses of psychoses
  with agreement about 90 to 94 . Validity is more
  difficult to verify.
• Diagnoses of psychiatric disorders rest on
  clinical interview and diagnostic criteria.
  (Lurie, 1992).

21
Prevalence of Psychiatric Disorders Among
National Health Insurance Enrollees in Taiwan

• Published paper
• Chien IC, Chou YJ, Lin CH, Bin SH, and Chou P
  Prevalence of Psychiatric Disorders Among
  National Health Insurance Enrollees in Taiwan.
  Psychiatr Serv. 55(6)691-697, 2004

22
The aim of this study

• Investigate the prevalence rates of psychiatric
  disorders among NHI members
• Discuss factors associated with prevalence rates
  of psychiatric disorders
• Compare the prevalence rates of psychiatric
  disorders between community survey (TPEP) in 1985
  and the NHI study in 2000.

23
Material and Methods

• Study Flow Chart

24
Prevalence of Psychiatric Disorders among
National Health Insurance Enrollees in Taiwan
NHI data file
Random sampling
Sampling file N200,432
Exclude age lt18
Study sample N137,914
1.priority in management of more than one
diagnosis (1)CIR (2)Frequency of
admission (3)Frequency of outpatient
(4)Diagnosed by psychiatrist 2.manage A code
Any minor psychiatric disorder
Any major psychiatric disorder
Prevalence
Compare with Community Survey in 1985
Associated factors
Any major psychiatric disorder ICD-9CM290,291,29
2,293,294,295,296,297,298 Any minor psychiatric
disorder ICD-9CM300,300.0,300.01,300.01(except30
0.01),300.1,300.2,300.3,300.4,300.5 300.6,300.7,30
0.8,300.81,300.89,301,302,306,307,308,309,310,311,
316
25
Table2. Demographic characteristics of the study
sample
26
Table 3. Prevalence of psychiatric disorders by
sex
Plt0.05 Plt0.01 Plt0.001 Prevalence lt0.005
27
Table 3. Prevalence of psychiatric disorders by
sex (cont)
Plt0.05 Plt0.01 Plt0.001 Prevalence lt0.005
28
Table 3. Prevalence of psychiatric disorders by
sex (cont)
Plt0.05 Plt0.01 Plt0.001 Prevalence lt0.005
29
Table 4. Logistic regression analysis of the
associated factors for psychiatric disorders
Plt0.05 Plt0.01 Plt0.001
30
Table 4. Logistic regression analysis of the
associated factors for psychiatric
disorders (cont)
Plt0.05 Plt0.01 Plt0.001
31
Table 5. Comparisons of the prevalence between
the community survey(1985) and the
National Health Insurance study (2000) in Taiwan
Ratio lt0.005 excludes foreigners from
National Health Insurance (NHI) sample in this
comparison
32
Discussion

• No study has ever been done using the BNHI files
  to measure the prevalence of psychiatric
  disorders in Taiwan.
• No update has been undertaken of any
  epidemiological community study of psychiatric
  disorders after the 1985 study
• The one-year prevalence rate of any major, any
  minor, and any psychiatric disorders was 1.37,
  4.26, and 5.30, respectively. These rates are
  lower than those of previous community studies
  (Compton III, 1991 Hwu, 1989 Cheng, 1988).

33
Prevalence of major psychiatric disorders

• These results were affected by the original
  prevalence rates, service needs, and disease
  pattern of these diseases.
• However, all the treated prevalences of major
  psychiatric disorders are much lower than those
  of community survey except schizophrenic
  disorders (Compton III, 1991 Hwu, 1989 Cheng,
  1988).

34
High prevalence of anxiety disorder

• The high prevalence of anxiety disorder was also
  noted in both the previous community study as
  well as with ambulatory care psychiatric patients
  (Hwu, 1989), which deserves further attention in
  terms of the diagnosis and clinical assessment of
  anxiety disorders.
• Perhaps we overdiagnose anxiety disorders and
  underdiagnose other neurotic disorders.

35
Prevalence of minor psychiatric disorders

• For the purpose of detail classification of
  prevalence in minor psychiatric disorders,
  comprehensive community field survey is
  indicated.
• To summarize, many patients with minor
  psychiatric disorders came to ask for help with
  problems of depression, sleep, or psychological
  stress.

36
Sex

• Females had higher prevalence rates in major
  depressive disorder and several minor psychiatric
  disorders, which was consistent with previous
  community findings that females have a higher
  prevalence related to depression, anxiety, and
  psychophysiological disorder (Hwu, 1989 Cheng,
  1988).
• The consistent finding of female preponderance in
  minor psychiatric disorders has also been
  demonstrated among Western countries (Weissman,
  1977 Binder, 1981 Bebbington, 1981 Dilling,
  1984 Madianos, 1985).

37
Sex

• Males had higher prevalence rates in
  schizophrenic disorder and other organic
  psychotic conditions in the present study, which
  indicates that more male patients with such
  disorders sought medical treatment.
• Generally, schizophrenic disorder is equally
  prevalent in males and females (Kaplan, 1998).
• Males had higher prevalence rates in alcoholic
  psychoses and drug psychoses. This indicates that
  males usually have more severe substance abuse
  problems (Hwu, 1989 Cheng, 1988).

38
Substance abuse

• We could not examine the data for alcohol abuse,
  alcohol dependence, drug abuse, and drug
  dependence because the NHI in Taiwan does not
  reimburse patients for care of a substance use
  disorder unless they also suffer from a psychotic
  state, such as alcoholic psychoses or amphetamine
  psychoses.
• So it is worthwhile to conduct a study to
  determine whether this particular reimbursement
  policy is having an adverse effect on the
  prevention and treatment of substance use
  disorders in Taiwan.

39
Age

• The prevalence rates for both any major and any
  minor psychiatric disorders increased with age.
• This indicates that we need to emphasize the use
  of psychiatric surveys and mental health care for
  the elderly population.
• In some community study, the age group with
  highest rates for most psychiatric disorders was
  found to be young adults (aged 25 to 44 years)
  (Robins, 1984).

40
SES

• We could conclude from this result that people
  with a higher rate of major psychiatric disorders
  showed a positive correlation with a lower SES,
  but we could not necessarily make the
  differentiation in this study that a lower SES
  was either a cause or a result (Dohrenwend,1990).
• Moreover, the lowest SES (fixed premium) had a
  significant effect on the prevalence of minor
  psychiatric disorders, which revealed more
  psychosocial stress in this group.

41
Urbanicity

• There were no significant differences among the
  three types of psychiatric disorders in terms of
  urbanicity, which was not consistent with the
  findings of the previous study, in which people
  in small towns and rural villages had a higher
  prevalence rate of minor psychiatric disorders
  than those in metropolitan areas (Hwu, 1989).
• In the present study, it is possible that the
  partial data was a representation of the place
  the insurance transaction took place, not the
  actual place of residence for this study.

42
Nationality

• In our study, natives had higher prevalence rates
  for any major, any minor, and any psychiatric
  disorders than did foreigners. This result
  indicated that most persons who were not native
  to Taiwan may have received physical and mental
  examinations before they entered the country.
• A cultural gap may have also hindered persons who
  were not native to Taiwan from seeking
  psychiatric treatment.

43
Nationality

• Foreigners have occupied 3 of total NHI
  enrollees, which are new citizens of Taiwan. They
  contain technicians, labor, maid, and women
  married to Taiwan and they face some cross
  cultural problems to adjust.
• We must follow and emphasize the general and
  mental health care of these enrollees in NHI.

44
Comparison between TPEP in 1985 and the NHI study
in 2000

• The prevalence of schizophrenic disorder was
  higher in the NHI study (0.46) than in the
  community survey (0.27, or 0.31 if we include
  schizophreniform disorder).
• This reveals that most schizophrenic patients
  receive treatment in Taiwan.

45
Why was the treated prevalence in the NHI study
higher than that found in the community study?

• First, the community survey may have omitted
  those persons with schizophrenia who had been
  admitted to hospitals.
• Second, social stigma can lead to a denial of
  symptoms or underreporting of previous episodes,
  which results in underestimation (de Salvia,
  1993).

46
Why was the treated prevalence in the NHI study
higher than that found in the community study

• Third, the 15-year gap, different sampling
  methods, and different diagnostic criteria
  probably contributed to the differences.
• Fourth, the possibility of an overdiagnosis of
  schizophrenia in NHI data for insurance
  application should be taken into consideration.

47
Comparison between TPEP NHI

• Bipolar affective disorder showed an
  approximately equal prevalence in both studies,
  which seems to indicate that most patients with
  bipolar affective disorder had also received
  treatment.
• However, the other 8 psychiatric disorders had
  much higher prevalence rates in the community
  survey than in the NHI study.

48
Prevalence rates of psychiatric disorders

• We also found that the prevalence rates of
  psychiatric disorders in the previous community
  survey in Taiwan were generally lower than those
  found in other countries using the same study
  design during the same time period (Compton III,
  1991 Hwu, 1989).
• These results are possibly due to methodological,
  cultural, and social factors, as well as to
  cross-national differences.

49
Prevalence rates of major depressive disorders

• In particular, the prevalence of major depressive
  disorder is strikingly low in both the previous
  community survey (1.14) and our survey of
  treated prevalence (0.35).
• We concluded that major depressive disorders in
  Taiwan are both underdiagnosed and undertreated.

50
Health care expenditures

• Total health expenditures in Taiwan, at 6 of
  Taiwans gross domestic product, are low compared
  with Germany (10.7 of the gross domestic
  product) and the United States (13 of the gross
  domestic product) (OECD, 2002 Lu, 2001).
• The expenditure for psychiatric disorders as a
  proportion of health spending (3) is low
  compared with those in Western countries, which
  is in agreement with our studys finding of a low
  treated prevalence of psychiatric disorders.

51
Advantage

• Using insurance data had many advantages,
  including a large available number for the sample
  and the saving of time and money needed to
  perform psychiatric assessments.
• Our study used random sample to prevent
  selection bias.

52
Limitation

• However, we faced some limitations, such as the
  reliability and validity of the secondary data,
  and the occurrence of over- and underdiagnoses,
  dual diagnoses, and primary and secondary
  diagnoses (Lurie, 1992).
• Also, there is a 15-year gap between the
  community survey and the NHI study, which creates
  obstacles to making accurate comparisons.

53
Limitation

• It is inevitable to exist bias in the estimation
  of prevalence due to use of service utilization
  data as opposed to community surveys.
• In fact, our findings represent use,
  availability, and accessibility of services
  rather than actual rates of disorders.
• We must also consider the implications of
  comparing rates with other studies, given
  different study designs and instruments.
• Surely, different instruments have different
  validity in making diagnoses.

54
Conclusions

• Both major and minor psychiatric disorders were
  undertreated in Taiwan.
• It is necessary for the public health department
  and the general populace to put more emphasis on
  mental illness education, prevention, and
  treatment
• In addition, we anticipate an upcoming
  epidemiological community survey that will be
  compared with the NHI data, to provide
  evidence-based implications for future policy
  making.

55
Future scope of study

• Disease type (Schizophrenia, Depression, Bipolar
  disorder et al)
• Comorbidity (Mental, General)
• Costs (Drug, Non-drug)
• Outcome evaluation
• Longitudinal follow-up

56
Accepted paper

• Date 09/29/04CC Subject Psychiatric Services
  Decision - APPI-PS-00331-2003.R2 "Health Care Use
  . . .Dear Dr. ChouWe are happy to advise you
  that the revised version of your paper entitled
  "Health Care Use and Costs of Psychiatric
  Disorders Among National Health Insurance
  Enrollees in Taiwan"(APPI-PS-00331-2003.R2) has
  been accepted for publication in Psychiatric
  Services.

57
????????

• ??
• ??
• ??

58
????

• ?????
• ????
• ??????

59
????????

• ??????,???????143?,???750?,????160?,??????6000??,?
  ????????????
• ????????
• ??30??????

60
??????????

• ????????????????????????lt???gt
• ?????????????????lt???gt
• ????B12???????????????,???????????????lt???gt
• ??????????????????lt???gt????
• ??????????????????lt???gt

61
??????????

• ????????????????????lt???gt
• ????????????-??-????????????????lt???gt
• ????Zolpidem ??????????????????lt???gt
• ????????????????????????lt???gt
• ????????????????????????????lt???gt
• ??????????????????????????lt???gt

62
??????????

• ??????????????lt???gt
• ??????????????lt???gt
• ??????????????????????(19962001?)lt???gt

63
93????????????

• ???????????????C??????????lt???gt
• ???????????????????????????? lt???gt
• ?????????????????????????????????lt???gt
• ???????????lt???gt

64
94????????????

• ???921?????????????????lt???gt
• ??????????????????????????????????????????lt???gt
• ???????lt???gt

65
????

• ????????????
• ???????

66
????????

• ??????
• ??????
• ??????
• ??????

67
??????????

• ????????????10?30?
• ???????
• ????????
• ?????????
• ??????
• ???????????

68
???????????

• ????
• ????
• ????
• ??
• ??
• ????
• ????
• ??

69
????

• ??
• ??
• ??
• ??
• ??
• ??
• ??
• ??

70
????

• ???????
• ??????
• ??????

71
??

• ??????????????????
• ?????????????????
• ????????????,??????????
• ?????????
• ?????????????

72
The End

• Thanks for your attention.