Physician contact by older Asian Americans: the effects of perceived mental health need.

OBJECTIVE: The use of physicians is more common than of behavioral specialists, especially in underserved Asian American communities. Despite a rapidly aging Asian American population, research has overlooked older people. This study examines the way mental health need affects the number of physician contacts by older Asian Americans.
METHOD: This study uses data on self-identified Asian Americans aged over age 50 years derived from the 2001 California Health Interview Survey. A total of 1191 Asian Americans from Chinese, Filipino, Korean, and Vietnamese backgrounds were studied. Replicate weights were applied to account for the survey's complex sampling methods. Linear regression was used to identify the number of physician contacts.
RESULTS: Overall, respondents had seen a doctor an average of five times in the previous 12 months; 7% perceived that they had a mental health need. Perceiving a mental health need was associated with a decreased number of physician contacts for Filipino and Korean Americans.
CONCLUSION: This study revealed interethnic differences among older Asian Americans' contact with physicians. As older Filipino and Korean Americans who perceive a mental health need have fewer contacts with their physician, correctly identifying mental health needs in the health care system for these groups is crucial. Health and mental health professionals can work toward reducing mental health disparities by accounting for older Asian Americans' help-seeking patterns when designing evidence-based interventions.

The increased use of physicians by those with mental health needs has been well documented.1 The use of physicians is more common than of behavioral specialists,2 especially in underserved Asian American communities.3 Despite a rapidly aging population, health research has overlooked older Asian Americans.4 In an effort to contribute to the knowledge base on a rapidly growing, yet understudied group, this study examines the way mental health need affects the frequency of contacts with a physician by older Asian Americans.

Prior research on services has focused on assessed need while overlooking the role of perceived mental health. Studies using the National Comorbidity Survey point to the role of perceived need in affecting mental health-related service studies.3,5 Research using Korean American elders describes the way perceived mental health need is shaped by cultural values,6 and this perception increases the number of physician contacts.

While focusing on a single ethnic group, previous research overlooks the diversity among Asian Americans.7,8 This study extends the existing literature by using the behavioral model9 to test the joint effects of the perception of mental health need on physician use among older Asian Americans. It is hypothesized that the number of physician contacts will differ by the joint effects of the perception of mental health need and Asian ethnic background.

Methods

Data source

This study uses data derived from the publicly available 2001 California Health Interview Survey (CHIS).10 The CHIS 2001 was selected for this study because of the availability of mental health need and service use variables that are not included in later versions of the California Health Interview Survey. The CHIS 2001 is a cross-sectional study of California residents’ health and access to health care services.11 A list-assisted random digit dial telephone survey, CHIS 2001 used a two-stage sampling procedure to generate representative estimates throughout the state. Separate surveys were administered for children and adults. Advertising outreach was conducted to enhance the participation rate, with a focus on linguistic and cultural minorities. While Chinese and Filipino respondents were recruited from traditional probability sampling methods, oversampling methods were used with Korean and Vietnamese subgroups to increase sample sizes and enhance estimate precision. The survey was translated and administered in several Asian languages including Mandarin, Korean, and Vietnamese.12 A total of 57,848 adults responded to the phone survey. The CHIS 2001 sample design documentation reported an overall 38% response rate.11

For this study, data on self-identified Asian Americans from Chinese, Filipino, Korean, and Vietnamese backgrounds aged over 50 years were extracted from the CHIS. To increase the study’s ability to make inferences about specific Asian ethnic groups, respondents from other Asian ethnic groups were excluded due to limited observations. This yielded a study sample of 1191 Asian Americans. The protocol for this study was deemed exempt from full review by the local institutional review board.

Measures

The dependent measure for this study was the number of physician contacts. All respondents to CHIS 2001 were asked, “How many times have you seen a doctor about your own health in the past 12 months?”

The study applied the behavioral model9 to organize covariates of physician visits. Predisposing characteristics were: age; Chinese, Filipino, Korean, and Vietnamese ethnicity; gender; and nativity, foreign or US born. Insurance status, English proficiency, and living arrangement were used as enabling covariates.

The study used perceived mental health, general health, and mental health specialist use as need factors. The General Health item from the Short Form-12 (SF-12) Health Survey13 was used as a measure of self-rated health. The Excellent, Very Good, and Good categories are very similar and were collapsed into one category as Good. The Fair/Poor responses were categorized as Fair. Two questions in CHIS 2001 related to mental health need and service use. The first asked for the respondents perceived mental health need: “During the past 12 months, did you think you needed help for emotional or mental health problems, such as feeling sad, blue, anxious or nervous?” The second asked, “Not counting overnight stays, emergency room visits, or visits for drug or alcohol problems, in the past 12 months, have you seen a psychiatrist, psychologist, social worker, or counselor for emotional or mental health problems?”

Analyses

Replicate weights were applied using SAS-callable SUDAAN 9.0.1 (RTI International, Research Triangle Park, NC) to account for the survey’s complex sampling methods. Chi-square tests were used to examine bivariate associations between categorical variables. Linear regression was used to model the quantity of physician visits. Main effect and interactive models were tested to examine the moderating effects of the perception of mental health need by ethnic group. While Chinese and Vietnamese differ in their migration experiences, many Vietnamese in the United States are ethnic Chinese, and both groups share a common belief system14 that is distinct from Koreans and Filipinos.15 Therefore, in an effort to focus on potential differences in cultural beliefs among the primary Asian ethnic groups, the Chinese and Vietnamese groups were combined as the reference category for the multivariate analysis.

Results

The weighted descriptive statistics are presented in Table 1. Overall, respondents had seen a doctor an average of 4.9 times in the previous 12 months; 7.0% perceived that they had a mental health need.

Table 1

Sample description

Variable	n	Weighted size	Weighted percentage
Chinese	379	260,212	39.0
Filipino	272	249,362	36.0
Korean	247	82,420	12.3
Vietnamese	293	84,452	12.7
Perceived mental health need	116	46,940	7.0
Did not perceive mental health need	1075	620,507	93.0
Saw a mental health specialist	37	12,888	1.9
Did not see a mental health specialist	1154	654,559	98.1
Age (M, SD)	63.0 (8.97)
Physician contacts (M, SD)	4.9 (12.4)

Abbreviations: M, mean; SD, standard deviation.

The bivariate associations between the independent variables and the four Asian ethnic groups are presented in Table 2. The number of physician contacts ranged from a low of 4.1 for Chinese Americans to a high of 5.8 for Vietnamese Americans. One in five Vietnamese Americans perceived a mental health need, which was three times the frequency of any other group. Differences in the percentages of fair health status were also noted, 89.2% of older Vietnamese Americans were in fair health, compared with 72.1% of Korean Americans and 61.2% of Chinese and Filipino Americans. Among enabling factors, a high discrepancy of uninsurance was observed among the Asian ethnic groups; one in three older Korean Americans were uninsured, more than twice the frequency of the other three groups. A higher proportion of Chinese Americans were aged over 65 years compared with 38% of Filipino Americans and approximately 31% of Korean and Vietnamese Americans.

Table 2

Group-specific weighted bivariate analyses from the 2001 California Health Interview Survey

Characteristic	Chinese	Filipino	Korean	Vietnamese
Physician contacts (M, SD)	4.1 (5.6)	5.5 (16.5)	4.5 (8.3)	5.8 (6.2)
Age (M, SD)	63.4 (9.5)	63.5 (10.2)	62.0 (7.5)	61.2 (8.4)
English proficiency (M, SD)	1.7 (1.4)	2.5 (1.3)	1.5 (.8)	1.3 (.85)
Over 65 years of age (%)a	43.7	38.1	30.8	31.4
Male	46.1	42.3	42.9	45.1
Uninsureda	8.9	11.3	34.7	16.3
Living with othersb	90.7	93.2	86.2	94.6
Foreign borna	87.2	95.4	98.3	100.0
Fair healtha	61.2	61.2	72.1	89.2
Perceived mental healtha	4.2	6.0	5.6	20.0
Mental health specialist useb	0.9	2.0	1.1	5.8

Notes:

P < 0.0001;

P < 0.05.

Abbreviations: M, mean; SD, standard deviation.

The results of the linear regression model testing the moderating effects that physician visits had on older Asian Americans’ perception of mental health need are presented in Table 3. The results (F(df = 13) = 12.49, P < 0.0001) provide support for the study’s test hypothesis that the perception of mental health need and Asian ethnic background have joint effects. Controlling for covariates, Filipino and Korean Americans who perceived a mental health need had fewer contacts with a physician than the reference category. On average, older Filipino Americans had 3.7 (standard error [SE] = 1.38) fewer contacts, and Korean Americans had 3.5 (SE = 1.50) fewer contacts. In addition, the uninsured had 2.96 (SE = 0.98) fewer contacts with a physician compared with older adults with insurance. Meanwhile older Asian Americans in fair health had 2.47 (SE = 0.60) fewer physician contacts than those in good health.

Table 3

Linear regression results for the number of physician contacts

Variable	b (SE)	t-test	P-value
Intercept	9.67 (3.23)	3.00	0.0036
Male gender	0.86 (0.82)	1.06	0.2942
Over 65 years of age	−0.53 (0.97)	0.55	0.5833
Asian ethnic group
Filipino	1.55 (1.76)	0.88	0.3825
Korean	0.87 (0.67)	1.31	0.1937
(reference: Chinese/Vietnamese)
Uninsured	−2.96 (0.98)	−3.03	0.0033
Living alone	−0.61 (0.54)	−1.13	0.2636
Foreign born	−2.44 (1.70)	−1.44	0.1544
English proficiency	−1.30 (0.86)	−1.52	0.1332
Fair health	−2.47 (0.60)	−4.13	0.0001
Saw mental health professional	0.54 (0.79)	0.68	0.5013
Perceived mental health need	2.61 (.73)	3.59	0.0006
Perceived mental health need-Asian ethnicity interaction
Filipino, perceived mental health need	−3.76 (1.38)	−2.72	0.0081
Korean, perceived mental health need	−3.54 (1.50)	−2.36	0.0208
(reference: Chinese/Vietnamese, No perceived mental health need)

Abbreviation: SE, standard error.

Discussion

This study revealed differences in how the perception of mental health need affects the number of older Asian Americans’ physician contacts. Perceiving a mental health need reduces the number of physician visits for Filipino and Korean Americans. Past research using a regional sample compared with non-Hispanic Whites suggests that older Korean Americans’ increased physician use is associated with the presence of perceived physical and mental health conditions6 yet the current findings using state-level data indicate that the perception of mental health problems decreases the number of physician visits.

The findings draw attention to the unique help-seeking behaviors of older Filipino and Korean Americans. Lower rates of physician contacts may mean accumulating unmet physical needs that may exacerbate mental health concerns. Furthermore, fewer contacts with a physician present fewer opportunities to identify and address mental health concerns. Pescosolido and Boyer’s network episodic model16 asserts that the type of mental health service used depends on a variety of social factors. Rather than seeing a physician for their mental health needs, older Filipino and Korean Americans may seek care and support from other sources, if at all. Future research should examine where older Filipino and Korean Americans receive care, and what culturally congruent models of physical and mental health care would be the most appropriate.

Use of the CHIS 2001 poses some limitations. While the data include mental health-related measures, the data were collected nearly a decade ago, which may limit the inferences for an evolving health care system. Second, the existing data focus primarily on physical health conditions, so information on specific mental health conditions is not available. Finally, the small raw sample sizes for Asian Americans result in less precise variance estimation, which may result in an underestimate of differences.

Despite limitations, this study makes an important contribution to the research by the differences among older Asian American groups. Knowing how the perception of mental health need affects general physician use allows health and mental health professionals to access to mental health care. As mental health needs do not increase physician contact for all Asian American groups, the identification and treatment of mental health concerns need to be adapted for different groups. By tailoring intervention efforts to the help-seeking patterns of Asian American groups, mental health professionals can work toward reducing mental health disparities.