Predicting Help-Seeking Attitudes Toward Mental Health Services Among American Indian Older Adults

Abstract

American Indian (AI) older adults are vulnerable to mental health disparities, yet very little is known about the factors associated with help-seeking for mental health services among them. The purpose of this study was to investigate the utility of Andersen’s Behavioral Model in explaining AI older adults’ help-seeking attitudes toward professional mental health services. Hierarchical regression analysis was used to examine predisposing, enabling, and need variables as predictors of help-seeking attitudes toward mental health services in a sample of 233 AI older adults from the Midwest. The model was found to have limited utility in the context of older AI help-seeking attitudes, as the proportion of explained variance was low. Gender, perceived stigma, social support, and physical health were significant predictors, whereas age, perceived mental health, and health insurance were not.

Keywords

American Indian older adults help-seeking mental health service utilization

The mental health service needs of American Indian and Alaska Native (AI/AN) populations have drawn national attention, particularly given their experiences of pervasive mental health disparities (Gone & Trimble, 2012; Manson, 2000). Mental health has been identified as a top health problem confronting AI/AN populations, and it reportedly contributes more than one third of the demand for services (Manson, 2000). On average, AI/AN populations experience serious psychological distress at 1.5 times the rate of the general population (American Psychological Association [APA], 2010); disproportionate rates of depression, substance abuse and dependence, posttraumatic stress disorder (PTSD), and suicide (APA, 2010; Gone & Trimble, 2012; Sarche & Spicer, 2008) have been reported. AI/AN populations have a 65% higher suicide rate and a 4.2 lower life expectancy than the general population (Indian Health Service, 2014). Despite these disparities, AI/AN mental health needs are often ignored and overlooked (Gone & Trimble, 2012), and very little is known about factors associated with their help-seeking attitudes for mental health services. For the purpose of this research, help-seeking attitudes indicate the propensity of individuals to seek professional services for their mental health needs.

There is great diversity and variability within and across AI/AN populations. AI/ANs represent more than 5 million people (U.S. Census, 2010) in 566 federally recognized tribes (Indian Health Service, 2014) and approximately 400 non-federally recognized tribes. While age-related differences in mental health help-seeking attitudes and behaviors are well supported (Babitsch, Gohl, & von Lengerke, 2012), research on this aspect of variability among AI/AN populations is non-existent. In particular, there is a critical need to understand the mental health help-seeking attitudes of AI/AN older adults, especially as their numbers are projected to double between 2012 and 2060.

Andersen’s model of health services use is one of the most highly utilized help-seeking research models, especially for predicting help-seeking attitudes and behaviors within minority populations (Andersen, Davidson, & Baumeister, 2013; Jang, Chiriboga, & Okazaki, 2009; Jang, Kim, Hansen, & Chiriboga, 2007; Snowden & Yamada, 2005). Help-seeking attitudes are thought to be critical determinants of help-seeking behaviors (Fischer & Farina, 1995; Jang et al., 2007; Tijhuis, Peters, & Foets, 1990), and other research with older minority populations has used attitudes toward help-seeking as outcome indicators in Andersen’s model (Jang et al., 2009; Jang et al., 2007). Given that help-seeking attitudes are important precursors toward actual mental health help-seeking behaviors, the purpose of this study was to investigate the utility of Andersen’s Behavioral Model for understanding AI older adults’ help-seeking attitudes toward mental health services.

Help-Seeking Among AI/ANs

Culturally related factors have been found to be salient to help-seeking for mental health services among AIs (Hartmann & Gone, 2013). Very few AIs reportedly seek professional help for mental health, and many drop out due to discomfort and cultural incongruence between providers and clients, and because of the limited treatment methods available to them (Hartmann & Gone, 2013). For instance, ambivalence toward conventional mental health services may arise due to divergent worldviews and healing traditions (Hartmann & Gone, 2012). Some AI/AN people view formal services with mistrust due to the relationship of these services with assimilative tactics associated with colonization, which aimed to strip AIs of their cultural traditions and identities (Hartmann & Gone, 2012).

Furthermore, the frequent social distance in experiences between AI/AN clients and service providers, who are often from distinct ethnic identities and social backgrounds, can complicate building rapport and create discomfort among those seeking services (Hartmann & Gone, 2012). In a mental health needs survey of 374 adult AI Coloradans from 17 to 71 years of age, less than half of those experiencing mental health problems sought help for those problems (King, 1999). The foremost reasons given for non-utilization included self-reliance, a lack of knowledge about services, negative beliefs and distrust about the helping system, believing that services would not help, and financial constraints (King, 1999).

Despite these findings, help-seeking rates have been found to vary significantly by gender, tribe, and region. In Gone and Trimble’s (2012) review of the extent research on mental health service utilization, a notable trend across tribes in different regions was that many AIs used mental health services, including biomedical services. However, many AIs have been found to prefer traditional healing methods (consulting a traditional healer, offering tobacco and praying, and/or participating in a sweat lodge) to conventional biomedical services (Beals et al., 2005; Gone & Trimble, 2012; Walls, Johnson, Whitbeck, & Hoyt, 2006).

In one of the reviewed studies (Beals et al., 2005), help-seeking for mental health services was relatively high among a large randomized sample of Southwestern and Northern Plains AI/ANs aged 15 to 54, and in some cases it was higher than among the general population. Preference for traditional healers depended on the type of problem for which AIs sought help (substance use disorders, anxiety and depressive disorders, suicide, and co-occurring disorders) and was quite common across populations, especially in the Southwest. In a sample of AI/ANs in Colorado, culturally sensitive and traditional services were the mental health services most preferred (58%), but the majority of participants were open to formal services (King, 1999). Likewise, among 865 parents/caretakers of tribally enrolled youth in the Northern Midwest, participants strongly preferred traditional informal services over biomedical services (Walls et al., 2006). Walls et al. (2006) recommended integrating traditional informal services into the biomedical model. Given the variability of help-seeking factors by tribe, region, and help-seeking beliefs, the need for localized understandings of help-seeking attitudes is apparent.

Andersen’s Behavioral Model

Andersen’s model of health services use, one of the most extensively researched help-seeking models, includes predisposing, need for service, and enabling explanatory factors (Andersen et al., 2013; Snowden & Yamada, 2005), and it incorporates individual and contextual factors related to utilization of health services (Babitsch et al., 2012). Predisposing factors are existing conditions (e.g., age, gender) that may predispose people to use or not use services, though they are not directly responsible for use; need factors (e.g., perceived mental health) indicate the need for services (Andersen et al., 2013); and enabling factors (e.g., social support, perceived stigma) facilitate or impede service use. In a review of 16 studies using Andersen’s model, the majority of studies used age and gender as predisposing factors, self-perceived health as a need factor, and health insurance and income/financial as enabling factors (Babitsch et al., 2012).

The extensive application of Andersen’s model in research indicates its recognized utility for predicting help-seeking attitudes and behaviors. Indeed, Babitsch et al. (2012) conducted a systematic review of all (N = 15) articles that applied Andersen’s model between 1998 and 2011 and found significant support for the model across these studies. Using Andersen’s model, researchers have found that health service utilization is socially patterned (Dhingra, Zack, Strine, Pearson, & Balluz, 2010). For example, non-Whites, and the poor, uninsured, and elderly were shown to have a lower likelihood of service use (Stockdale, Tang, Zhang, Belin, & Wells, 2007).

Although Andersen’s model has been used to study ethnic minorities, it has not been applied to AI populations, including AI older adults. Whether Andersen’s model is a good fit for research on AI populations is an important consideration in itself. Andersen’s model does not account for divergent worldviews and conceptualizations of mental health, which applies to many AIs (Hartmann & Gone, 2012, 2013). Andersen’s model also does not account for preferences for traditional health methods, and many AIs prefer traditional healing over formal treatments (Beals et al., 2005). Thus, an important question for research is whether Andersen’s model is a good fit for research on AI older adults.

Predisposing Factors: Age and Gender

Age and gender are common predisposing factors used within Andersen’s framework (Jang et al., 2009; Jang et al., 2007; Lindamer et al., 2012; Snowden & Yamada, 2005). Older adults are thought to utilize mental health services less than their younger counterparts, and women are thought to use them more than men (Cauce et al., 2002; Jang et al., 2007). In a study of 92,362 Canadian adults, those who were 75 years of age and older had the lowest odds of seeking specialist consultations. But in another study (Bazargan, Bazargan-Hejazi, Andersen, Hindman, & Baker, 2008), African American and Latino participants over 60 were more likely to have obtained a physician diagnosis for their medical conditions than younger counterparts. Other studies have identified a trend of older minority adults being more likely to seek services for mental health needs than younger minority adults (Babitsch et al., 2012).

Need for Service Factor: Perceived Mental Health

Perceived health and mental health are commonly indicative of need within Andersen’s model (Babitsch et al., 2012; Dhingra et al., 2010; Snowden & Yamada, 2005). Problem recognition is closely linked to use of services, yet research indicates an inconsistent relationship between need and help-seeking (Cauce et al., 2002). Although research on help-seeking among AI older adults is lacking, research with 64 AI adolescents found no significant relationship between symptoms of psychopathology and concomitant receipt of mental health treatment (Novins, Beals, Shore, & Manson, 1996). Another study of 582 AIs in the Southwest found strong relationships between the need for services, as indicated by the number of psychiatric diagnoses, and likelihood of receiving mental health treatment (Robin, Chester, Rasmussen, Jaranson, & Goldman, 1997).

Enabling Factors: Stigma, Social Support, and Health Insurance

Stigma, social support, and health insurance can either deter or facilitate mental health service utilization, depending on the context. Several studies have examined stigma and other cultural factors in relationship to AI help-seeking behaviors. In a sample of 101 AI/ANs in the Southwest aged 15 to 21, stigma about receiving services, lack of problem recognition, and the belief that no one could help were barriers to help-seeking for suicide symptomatology (Freedenthal & Stiffman, 2007). Focus groups with 107 AI urban youths and their families found stigma around services to be a barrier to accessing service; other barriers were lack of trust in services and a preference for using community supports and family members instead of professional services (West, Williams, Suzukovich, Strangeman, & Novins, 2012).

Social support is often considered an enabling factor that can either prevent or bolster help-seeking behavior (Cauce et al., 2002; Dhingra et al., 2010; Snowden & Yamada, 2005). AIs most often seek help for mental health concerns from informal helpers such as family and friends (Freedenthal & Stiffman, 2007). Some AIs embrace self-reliance to get through personal troubles (Freedenthal & Stiffman, 2007), and the type and quality of social support they receive can significantly affect their help-seeking behaviors. In a study of 3,084 AIs aged 15 to 54, those who received instrumental social support and counseling reported self-reliance to be less of an obstacle to help-seeking than those who received negative social support characterized by criticism (Duran et al., 2005).

A lack of insurance is a common barrier emphasized in research on help-seeking among minority populations (Snowden & Yamada, 2005). Within Andersen’s framework, health insurance is often considered an enabling factor (Dhingra et al., 2010; Jang et al., 2009). Using a nationally representative sample of approximately 2,500 AIs younger than 65, those with less insurance coverage had worse access and health care utilization than similar Whites (Zuckerman, Haley, Roubideaux, & Lillie-Blanton, 2004). Research has not examined insurance in relationship to AI older adults’ help-seeking behaviors or attitudes.

The current study aimed to examine the usefulness of Andersen’s Behavioral Model for understanding AI older adults’ attitudes toward help-seeking for mental health services. More specifically, the study examined how well predisposing, need, and enabling factors predict help-seeking attitudes.

Method

Sample and Data Collection

Using a cross-sectional design, data were collected from a convenience sample of rural and off-reservation self-identified AI older adult participants, aged 50 or older, from January to May 2013. Rural was defined as areas with a population of fewer than 50,000 people (Hack, Larrison, & Gone, 2014; U.S. Census Bureau, 2011). The cutoff age of 50 was selected because of the reduced life expectancy of AI older adults in comparison with the general population (Indian Health Service, 2014), as has been done in previous research (e.g., Hendrix, 2010). Participants were recruited from a variety of off-reservation locations, including AI churches, social service centers, other religious organizations, senior housing facilities, senior centers, and three powwows in two Midwestern states. A total of 235 AI older adults participated in the study. Two participants who failed to complete the questionnaires were excluded, resulting in a final sample of 233 AI older adults.

While questionnaires were self-administered, trained AI interviewers were available to assist anyone who might need assistance. Two participants completed the questionnaire with assistance. Before administering the survey, interviewers explained the study purposes, the procedures, and the scope of questions that would be asked, confidentiality precautions, as well as the benefits and risks of the study. All participants gave informed written consent prior to interview. The questionnaire took approximately 30 min to complete, and participants were offered US$10 for their time. All procedures were approved by the University of South Dakota, Institutional Review Board.

Measures

Dependent variable

Attitudes toward seeking professional psychological help were measured with a 10-item version of the Attitudes Toward Seeking Professional Psychological Help (ATSPPH; Fischer & Farina, 1995) scale. This scale includes five positive statements (e.g., “If I believed I was having a mental breakdown, my first inclination would be to get professional attention”) and five negative statements (e.g., “The idea of talking about problems with a psychologist strikes me as a poor way to get rid of emotional conflicts”). Items were rated on a 4-point scale ranging from (0) disagree to (3) agree. Higher scores indicate more positive views toward seeking formal mental health services. A Cronbach’s alpha of .83 was reported for AI college students (Price & McNeill, 1992); alpha was .68 in the current study. This relatively low estimate of internal consistency was not anticipated, given the much higher alpha in its previous use. Differences in the samples likely account for the divergent estimates of internal consistency, especially differences in age and level of education.

Independent variables: Predisposing

Predisposing factors reflect propensity to use services independent of any personal circumstances or experiences that may cause the need for service use. In the current study, these were demographic and background characteristics, including age (a continuous variable), gender (female = 1, male = 0), and employment (employed = 1, not employed = 0).

Independent variables: Need for service

Participants’ overall perceived mental health condition was considered to represent need for service. A single item asked, “How would you rate your overall mental health at the present time?” A 4-point response ranging from (1) poor to (4) excellent was used for this item.

Independent variables: Enabling

Perceived stigma

Stigma toward receiving psychological help was measured with the Social Stigma for Receiving Psychological Help (SSRPH; Komiya, Good, & Sherrod, 2000) scale. The SSRPH was devised to evaluate individuals’ perceptions of how stigmatizing it is to receive professional psychological help. It consists of five items rated on a 4-point scale from (1) strongly disagree to (4) strongly agree. Scores from 12 to 20 indicate greater perception of social stigma associated with receiving professional psychological help; scores from 4 to 11 indicate less social stigma. A sample item is “people tend to like less those who are receiving professional psychological help.” Cronbach’s alpha was .72 among a sample of college students (Komiya et al., 2000) and .80 in the current study.

Perceived physical health

This factor was measured with a single item: “How would you rate your overall physical health at the present time?” Participants were asked to rate their health on a 4-point scale ranging from (1) poor to (4) excellent.

Social support

The Multidimensional Scale of Perceived Social Support (MSPSS; Zimet, Dahlem, Zimet, & Farley, 1988) was used to measure perceived global social support (i.e., from family, friends, and significant others). This instrument consists of 12 items (e.g., “there is a special person that is around when I am in need”) and a 4-point response format that ranged from (1) strongly disagree to (4) strongly agree. The scale has been used in several studies to measure global social support among various older adult populations, including those in rural areas, immigrant older adults, and residents in assisted living facilities (Cummings & Cockerham, 2004; Lee & Woo, 2013). Higher scores reflect higher levels of perceived social support. Cronbach’s alpha was .94 in the current study.

Health insurance

A single item using a categorical (yes = 1, no = 0) response format asked whether participants had health insurance coverage.

Analytic Strategy

Hierarchical linear regression was used to examine predictors of attitudes toward seeking professional psychological help among AI older adults. Predisposing variables (age, female, and being employed) comprised the first set of predictors, followed by the need for service variable (perceived mental health). Enabling variables (stigma, perceived physical health, social support, and health insurance) were added in the third step. IBM SPSS version 21 (IBM, 2012) was used to conduct the analysis. There were no multicollinearity problems among the predictors, as indicated by Variance Inflation Factor (VIF) scores that were all greater than 1.012 (Mertler & Vannatta, 2002). In addition, all variables were normally distributed except age, for which a square root transformation was used. For each predictor, the unstandardized regression coefficient represented the relationship with the outcome variable while controlling for the other predictors.

Results

Description of the Study Participants

As shown in Table 1, the age of participants ranged from 50 to 95, with a mean of 60.7 years. A little more than half were female, and just over one third were married. Most participants had at least a high school degree/General Education Development (GED). About one half of the participants were employed. Among the participants, 71% reported good or excellent physical health, and 83.4% reported good or excellent mental health. The characteristics of this sample were similar to those reported for AIs in a previous national study (Goins & Pilkerton, 2010).

Table 1.

Descriptive Characteristics of Study Participants (N = 233).

	n	% or M
Age
Range = 50–95		60.7
Gender (n = 230)
Female	125	54.3
Male	105	45.7
Marital status (n = 231)
Married	84	36.4
Never married	41	17.7
Others	106	45.9
Education (n = 227)
Lower than high school diploma/GED	19	8.3
High school diploma/GED or greater	208	91.7
Being employed (n = 224)
Yes	110	49.1
No	114	50.9
Health insurance (n = 231)
Yes	145	62.8
No	86	37.2
Perceived physical health (n = 231)
Poor or fair	67	29.0
Good or excellent	164	71.0
Perceived mental health (n = 199)
Poor or fair	33	16.6
Good or excellent	166	83.4

Note. GED = General Education Development.

Descriptive Analyses

Table 2 shows the mean scores, ranges, standard deviations, and correlations for the main variables. None of the variables had highly skewed distributions. Attitudes toward seeking professional psychological help were positively correlated with social support (r = .22, p ≤ .01) and negatively correlated with stigma toward receiving psychological help (r = −.21, p ≤ .01). Neither perceived mental health nor perceived physical health correlated significantly with attitudes toward seeking professional psychological help.

Table 2.

Pearson Correlations, Ranges, Means, and Standard Deviations Among Main Variables (N = 233).

	ATSPPH	Range	M	SD
ATSPPH		0-30	18.63	5.64
Perceived mental health	−.07	1-4	3.18	0.73
SSRPH	−.21**	5-19	10.52	3.18
Perceived physical health	−.06	1-4	2.83	0.72
Social support	.22**	12-48	38.69	6.89

Note. ATSPPH = Attitudes Toward Seeking Professional Psychological Help; SSRPH = Stigma Toward Receiving Psychological Help.

p ≤ .01.

Predictors of Attitudes Toward Seeking Professional Psychological Help

The hierarchical regression results on predictors of attitudes toward seeking professional psychological help are shown in Table 3. In Step 1, the predisposing variables explained 6.8% of the variance in attitudes toward seeking professional psychological help. Female gender was a significant predictor of positive attitudes (B = 2.553, p ≤ .01). The need for service variable of perceived mental health was added in Step 2, but it added a negligible amount to the explained variance. Female gender remained a significant predictor of positive in Step 2 (B = 2.521, p ≤ .01).

Table 3.

Hierarchical Regression Model of Attitudes Toward Seeking Professional Psychological Help (N = 233).

	Attitudes toward seeking professional psychological help
	B^a (SE)
	Step 1	Step 2	Step 3
Predisposing variables
Age	−0.976 (0.815)	−1.007 (0.819)	−0.822 (0.828)
Female	2.558 (0.855)**	2.527 (0.860)**	1.719 (0.860)*
Being employed	−1.198 (0.878)	−1.142 (0.888)	−0.817 (0.884)
Need variable
Perceived mental health		−0.280 (0.577)	−0.854 (0.655)
R² change at this step		.001
Enabling variables
Stigma toward receiving psychological help			−0.361 (0.139)**
Perceived physical health			−1.359 (0.644)*
Social support			0.186 (0.070)**
Health insurance			0.580 (0.898)
R² change at this step			.085
F	3.755*	2.861*	3.440***
R² total	.068	.069	.154

Note. Standard errors are given in parentheses.

Unstandardized coefficients.

p ≤ .05. **p ≤ .01. ***p ≤ .001.

In the final step, the enabling variables added an additional 8.5% to the explained variance. The full model accounted for 15.4% of the variance in attitudes toward seeking professional psychological help. Higher levels of stigma toward receiving psychological help (B = −0.361, p ≤ .01) and higher levels of perceived physical health (B = −1.359, p ≤ .05) predicted more negative attitudes toward seeking professional psychological help. By contrast, higher levels of perceived social support predicted more positive attitudes toward seeking professional psychological help (B = 0.186, p ≤ .01), as did female gender (B = 1.719, p ≤ .05). As indicated by the regression coefficients, a 1-point increase in the score on perceived physical health decreased the predicted score on attitudes by more than one and a half points; a 1-point increase in stigma decreased the predicted score by 0.361 points. Conversely, a 1-point increase in social support increased the predicted score on attitudes by 0.186 points. Age, employment, health insurance, and perceived mental health were not significant predictors.

Discussion

The study’s results partially supported Andersen’s Behavioral Model, in that predisposing and enabling factors, but not need for service, predicted attitudes toward receiving professional psychological help. Prior to discussing these findings in greater detail, we will turn our attention to certain descriptive statistics that may shed some light on the findings. First, the mean score on attitudes toward mental health services (18.63) was quite a bit higher than the mean score (13.00) that Jang et al. (2007) reported for a sample of older Korean American adults; these scores indicate slightly positive attitudes toward mental health services among older AIs in comparison with slightly negative attitudes among older Koreans. Reports on help-seeking behaviors among AIs vary considerably; one study (Beals et al., 2005) reported that Southwest and Northern Plains tribal members aged 15 to 54 sought mental health assistance at a greater rate than non-AIs, but other research reported that AIs tend to underutilize conventional biomedical mental health care services (Hartmann & Gone, 2012). Thus, there appears to be variability in the help-seeking attitudes and behaviors of older AIs depending on sample and setting.

Second, the current sample’s mean score on perceived stigma toward receiving psychological services was relatively low, which is important given stigma is a significant barrier to help-seeking (Freedenthal & Stiffman, 2007; West et al., 2012; Westermeyer, Canive, Thuras, Chesness, & Thompson, 2002). Being recruited from community events or agencies, the current sample may perceive less stigma toward accessing psychological services than older AIs who do not access these resources.

Third, perceived social support was quite high in the current sample, which is consistent with the tendency of AIs to rely on family and friends for support (Freedenthal & Stiffman, 2007). Indeed, connectedness and interdependency with family and friends are paramount among AI communities (Duran et al., 2005). However, age-related losses such as death of loved ones, reduced social contact due to retirement, and social isolation due to functional impairment can increase with age, and loneliness and limited social support are risk factors for mental health problems (Singh & Misra, 2009). Thus, the sample’s perceived high social support may help to explain its scores on perceived mental health, which were also rather high.

Predisposing Variables

The predisposing variable of gender was predictive of help-seeking attitudes toward professional psychological help, with women more likely to have positive views toward seeking services than men. More specifically, female gender predicted scores on attitudes more than one and a half points higher than male gender. This finding is consistent with that of Jang et al. (2007) as well as the majority of research using Andersen’s model (Babitsch et al., 2012). Moreover, self-reliance has been a salient value and norm among AI populations (Freedenthal & Stiffman, 2007; King, 1999), and it is plausible that self-reliance may be higher among AI males than females. Therefore, self-reliance might prevent males from seeking formal help to a greater degree than it would among females.

This study did not find age to be related to attitudes toward help-seeking. Given that the current study examined older adults exclusively, this finding indicates that attitudes toward receiving professional psychological help among AI adults in their 80s and 90s are similar to those of AI adults in their 50s and 60s, even though across a broader age range attitudes might vary.

Need for Service Variable

Perceived mental health, the need variable, failed to predict attitudes toward professional psychological services in the current sample. Previous studies examining the role of mental health status as an indication of need for service among younger AIs (Novins et al., 1996; Robin et al., 1997) have yielded mixed results. Similarly, the current research suggests that older AIs’ perceived or actual mental health status may not play a role in whether they seek professional mental health services. Older AIs might not recognize that they have a mental health problem, or they might perceive that the problem will subside on its own (Cauce et al., 2002). Given that physical health status predicted help-seeking attitudes better than mental health status did, it is possible that expressing mental health distress is culturally taboo in this population and the expression of distress through physical means is more acceptable. Moreover, older AIs may conceptualize mental health in culturally distinct ways, subscribing to holistic ideas of wellness, or a balance between the mind, body, soul, and spirit, rather than mental health (University of Maryland Center for School Mental Health, 2011). As an implication, a mismatch between mainstream and AI conceptions of mental health may act as a barrier to receipt of needed services.

Enabling Variables

The enabling variables of perceived physical health, stigma toward psychological help, and social support were predictive of help-seeking among AI older adults. Upon first analysis, the finding on perceived physical health appears counterintuitive, as better physical health may translate to greater ability to access services. However, perceived better physical health may also enable AIs to be more self-reliant, which could decrease their desire for formal services. Notably, self-reliance was one of the reasons given for non-utilization of mental health services in a previous study of both younger and older AI adults (King, 1999). Among older AIs, self-reliance has been associated with ethnic pride, which has been found to be a protective factor among some AI populations (Kulis, Napoli, & Marsiglia, 2002). In this way, self-reliance may deter formal help-seeking and still protect against mental health problems themselves.

As other studies (Freedenthal & Stiffman, 2007; West et al., 2012; Westermeyer et al., 2002) have also found stigma to predict less positive views toward mental health services among AIs, the current finding on this variable underscores the importance of recognizing its deterrent effect on the utilization of mental health services among these populations, including older AIs. The negative legacy of historical oppression of AIs (Burnette & Sanders, 2014) and concomitant mistrust of formal services may add to stigma in presenting significant barriers to mental health care utilization and health equity among AI populations (Hartmann & Gone, 2012).

Similar to other studies (e.g., Freedenthal & Stiffman, 2007), social support was predictive of more positive help-seeking attitudes for mental health services. This finding lends additional credence to the finding in other AI research (Duran et al., 2005) that friends and family bolstered help-seeking attitudes, rather than serving as a replacement for the services themselves. This finding suggests that social support can be an important facilitating factor related to help-seeking attitudes—one that may be martialed to complement and support formal mental health services.

Overall Fit of the Andersen Model

Because the model explained only about 15% of the variance in attitudes toward receiving professional help, it is clear that these attitudes are also associated with other variables not examined in this study. An important question to consider is whether these are variables that fit within Andersen’s model, or if an entirely different model should be used to understand help-seeking attitudes among older AIs. Another important question is how well Andersen’s model explains help-seeking attitudes, as compared with help-seeking behaviors, in this population.

Level of education, a predisposing factor not included in the current study because of a lack of variability in the sample, has been important in other research (Babitsch et al., 2012), and it might add explanatory value. Perceived cultural sensitivity of provider services, which would be considered an enabling factor, may also affect help-seeking attitudes of AI older adults (Gone & Trimble, 2012). Culturally sensitive provider services may be tribally based, match help seekers’ ethnic background or provide services in preferred languages, and address structural barriers to seeking services, such as providing transportation or in-home services. Traditional healing is often preferred by AI community members (Beals et al., 2005), and understanding the role of these and other informal services may explain additional variance in help-seeking attitudes toward formal psychological services. Although perceived need for mental health services did not predict help-seeking attitudes in the current study, alternative conceptualizations of need for services that incorporate holistic frameworks such as wellness might yield better results (West et al., 2012).

The current findings regarding the fit of Andersen’s model are consistent with similar studies doing so with Korean Americans; the model accounted for approximately 13% of the variance related to help-seeking attitudes of older Korean Americans (Jang et al., 2007) and 16% of the variance among another sample of Korean Americans (Jang et al., 2009). However, studies using Andersen’s model to explain help-seeking behaviors among Asian populations have yielded results that are quite different. For example, an adapted version of Andersen’s model accounted for approximately 36% of the variance in health service utilization among South Asian Canadians (Surood & Lai, 2010). Thus, modifications to the model seem warranted if it is to be used to explain help-seeking attitudes among AI older adults.

Limitations of the Study

Several limitations of the current study should be noted. The cross-sectional research design limits our ability to make causal conclusions about the findings. In addition, the sample is not representative of AI older adults throughout the United States. The use of a convenience sampling method to recruit AI older adults in two Midwestern states limits the generalizability of the findings to older AI in other settings or states. Because AI ethnicity was self-identified, it is possible that some participants might not be considered AI if verification through tribal membership were used to identify the sample. In addition, as data on tribal membership were not collected, we could not examine tribal differences in any of the examined variables. Selection biases might have affected the findings in several ways. Participants who chose to participate in the study might have been more willing to discuss the seeking of professional psychological help than those who did not choose to participate. They also might have held more positive views about mental health services, and they might have had fewer emotional or psychological problems. The level of help-seeking for professional psychological services among older AIs who are homebound or institutionalized might be different than those who are actively involved in senior centers and powwows who participated in this study. Studies with more representative samples of AI older adults generally and also across different tribes and rural/urban contexts will provide a fuller picture of physical and mental health effects, thereby advancing our gerontological knowledge base (Hack et al., 2014).

There are several limitations based on measurement choices used in the study. First, all of the data are based on self-report, and participants could have provided answers they considered to be socially desirable. Second, the low internal consistency of the outcome measure might have resulted in an attenuation of the relationships with the predictor variables, thereby reducing the model’s explanatory value. Additional development of this measure is warranted before it is used in further research with older AIs; strengthening its internal consistency as well as examining its content and criterion validity would be especially important. Third, the other study measures had not been used previously with older AIs. The measure of perceived mental health might not have captured adequately conceptualizations of mental health held by older AIs or the true level of mental health among them. Culturally grounded mental health or wellness instruments might better assess the relationship between mental health need and attitudes (Hendrix, 2010; Manson, 2000).

Implications of the Findings

The limitations notwithstanding, there are some important implications of the study’s findings. The study identified salient factors that should be taken into consideration when designing and adapting mental health services for older AIs, particularly stigma, social support, and perceived physical health. Given that stigma is a barrier to mental health help-seeking, addressing its causes represents an important first step. A reported underlying cause of stigma and mental health distress apparent in AI communities is historical oppression (Hartmann & Gone, 2013). Some AI people have found conventional mental health treatments with an individualistic focus to be oppressive in that they undermine tenets of their worldviews, including collectivism (Hartmann & Gone, 2013). Other empirical research with AI populations has emphasized stigma as a significant barrier to help-seeking behaviors and the importance of culturally relevant programs that promote wellness (Manson, 2000; West et al., 2012). For example, cultural insensitivity by providers and distrust in the mental health service system were barriers identified by AI veterans (Westermeyer et al., 2002). Thus, addressing stigma may mean addressing the cultural incongruence of some mental health services.

Results indicate that social support is clearly an important resource related to help-seeking among AI older adults. In a review of journal articles between the years 1980 and 2000, the importance of family and community in treatment and the role of traditional healing were salient factors related to mental health services for AI populations (Manson, 2000). Informal supports may provide an important alternative rather than a barrier to professional mental health services (Freedenthal & Stiffman, 2007). With many AI populations preferring traditional healing and the incorporation of friends and family members (Manson, 2000) into mental health interventions, these existing strengths within AI communities are important areas to inform treatment development. Given the importance of social support and AI preferences, family-focused interventions are viable alternatives to individually focused interventions. With the absence of highly trained AI mental health professionals (Hartmann & Gone, 2013), paraprofessionals may be necessary to meet the demand for mental health services. Thus, incorporating community members who have cultural knowledge of traditional healing can be a way to address the need for mental health professionals who can simultaneously inform culturally relevant mental health interventions.

Finally, with perceived physical health predicting mental health help-seeking attitudes of AI older adults, the connection between mental and physical health is apparent. The importance of medical providers assessing for mental health problems is crucial, and specific training for health professionals to identify mental health problems among these populations may be needed. Furthermore, a wellness model that incorporates mental, physical, and spiritual domains of health may serve to examine these connections holistically, in a culturally sensitive way.

Conclusion

In conclusion, Andersen’s Behavioral Model had limited usefulness in predicting help-seeking attitudes toward mental health services among older AIs. Because of this, more research is needed to understand the factors related to the mental health help-seeking attitudes of older AI adults. In particular, qualitative research with AI older adults may provide in-depth understanding of how AI older adults perceive both mental health and culturally sensitive services models to treat mental health problems. A culturally specific service model may then be developed inductively from these findings and examined across multiple AI older adult populations using quantitative research methods. Furthermore, research on the relationships among help-seeking attitudes and behaviors can delineate their precise connections.

Culturally grounded service models would likely include the incorporation and preferences for traditional healing, a relational worldview and a holistic perspective on mental health (such as wellness), informal supports, and acknowledgment of historical oppression and accounting for its effect on AIs’ potentially negative experiences with formal mental health services (Hartmann & Gone, 2012, 2013; King, 1999). As heterogeneity within and across AI populations is great, successful mental health service models must be sensitive to the variance across older AI service preferences and perspectives on mental health and avoid imposing a “one-size-fits-all” model.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Soonhee Roh, PhD, is an assistant professor at University of South Dakota’s Department of Social Work in Sioux Falls. Her areas of research interest include health disparities, positive adaptation in aging, mental health, as well as for access and behavioral service utilization issues among ethnic and racial minorities.

Catherine E. Burnette, PhD, is an assistant professor at Tulane University’s School of Social Work in New Orleans. Her research interests include indigenous issues, mental health, critical theory, historical oppression, family resilience, intimate partner violence and culturally relevant social work practice.

Kyoung Hag Lee, PhD, is an associate professor in the School of Social Work at Wichita State University. He is the director of the Social Work Research and Evaluation Lab. His research interests include the well-being of immigrant elders and assisted living residents.

Yeon-Shim Lee, PhD, is an associate Professor at San Francisco State University School of Social Work. She has published extensively on racial and ethnic health/mental health disparities, family violence (elder abuse and intimate partner abuse), substance use issues, acculturation, and health interventions.

James I. Martin, PhD, is an associate professor at New York University’s Silver School of Social Work. His research interests include stress and coping, constructions of identity among minority populations, and the impact of historical time on research findings.

Michael J. Lawler, PhD, is dean and professor, School of Health Sciences, University of South Dakota. His research interests include health disparities, child well-being, child welfare, transfer of learning, and forensic psychology.

References

American Psychological Association. (2010). APA fact sheet, mental health disparities: American Indian and Alaska Natives. Retrieved from http://www.psych.org/Share/OMNA/Mental-Health-Disparities-Fact-Sheet–American-Indians.aspx

Andersen

R. M.

Davidson

P. L.

Baumeister

S. E.

(2013). Improving access to care. In Kominski

E. F.

(Ed.), Changing the US health care system: Key issues in health services policy and management (pp. 33-69). San Francisco, CA: Jossey-Bass.

Babitsch

Gohl

von Lengerke

(2012). Re-revisiting Andersen’s behavioral model of health services use: A systematic review of studies from 1998-2011. GMS Psycho-Social-Medicine, 9, 1-15.

Bazargan

Bazargan-Hejazi

Andersen

R. M.

Hindman

D. W.

Baker

R. S.

(2008). Correlates of self-diagnosis of chronic medical and mental health conditions in under-served African American and Latino populations. Ethnicity & Disease, 18(2, Suppl. 2), S2-S105.

Beals

Manson

S. M.

Whitesell

N. R.

Spicer

Novins

D. K.

Mitchell

C. M.

(2005). Prevalence of DSM-IV disorders and attendant help-seeking in 2 American Indian reservation populations. Archives of General Psychiatry, 62, 99-108.

Burnette

C. E.

Sanders

(2014). Trust development in research with indigenous communities in the United States. The Qualitative Report, 19(44), 1-19.

Cauce

A. M.

Domenech-Rodríguez

Paradise

Cochran

B. N.

Shea

J. M.

Srebnik

Baydar

(2002). Cultural and contextual influences in mental health help-seeking: A focus on ethnic minority youth. Journal of Consulting and Clinical Psychology, 70, 44-55.

Cummings

S. M.

Cockerham

(2004). Depression and life satisfaction in assisted living residents: Impact of health and social support. Clinical Gerontologist, 27, 25-42.

Dhingra

Zack

Strine

Pearson

Balluz

(2010). Determining prevalence and correlates of psychiatric treatment with Andersen’s behavioral model of health services use. Psychiatric Services, 61, 524-528.

10.

Duran

Lucero

Jiang

Novins

D. K.

Manson

Beals

(2005). Obstacles for rural American Indians seeking alcohol, drug, or mental health treatment. Journal of Consulting and Clinical Psychology, 73, 819-829.

11.

Fischer

E. H.

Farina

(1995). Attitudes toward seeking professional psychological help: A shortened form and considerations for research. Journal of College Student Development, 36, 368-373.

12.

Freedenthal

Stiffman

A. R.

(2007). They might think I was crazy: Young American Indians’ reasons for not seeking help when suicidal. Journal of Adolescent Research, 22, 58-77.

13.

Goins

R. T.

Pilkerton

C. S.

(2010). Comorbidity among older American Indians: The Native elder care study. Journal of Cross-Cultural Gerontology, 25, 343-354.

14.

Gone

J. P.

Trimble

J. E.

(2012). American Indian and Alaska Native mental health: Diverse perspectives on enduring disparities. Annual Review of Clinical Psychology, 8, 131-160.

15.

Hack

S. M.

Larrison

C. R.

Gone

J. P.

(2014). American Indian identity in mental health services utilization data from a rural Midwestern sample. Cultural Diversity and Ethnic Minority Psychology, 20, 68-74.

16.

Hartmann

W. E.

Gone

J. P.

(2012). Incorporating traditional healing into an urban American Indian health organization: A case study of community member perspectives. Journal of Counseling Psychology, 59, 542-554.

17.

Hartmann

W. E.

Gone

J. P.

(2013). American Indian and Alaskan Native mental health. In Shally-Jensen

(Ed.), Mental health care issues in America: An encyclopedia (Vol. 1, pp. 40-47). Santa Barbara, CA: ABC-CLIO. Retrieved from http://gonetowar.com/wp-content/uploads/2013/11/alaskan-native-20130222.pdf

18.

Hendrix

(2010). Health and health care of American Indian older adults (Course Module No. 650-493-5000 x66206). Stanford, CA: eCampus Geriatrics, Stanford School of Medicine. Retrieved from http://web.stanford.edu/group/ethnoger/americanindian.html

19.

IBM. (2012). IBM SPSS Statistics for Windows (Version 21.0). Armonk, NY: Author.

20.

Indian Health Service. (2014). Indian health service fact sheets: Indian health disparities. Retrieved from http://www.ihs.gov/newsroom/factsheets/disparities/

21.

Jang

Chiriboga

D. A.

Okazaki

(2009). Attitudes toward mental health services: Age-group differences in Korean American adults. Aging and Mental Health, 13, 127-134.

22.

Jang

Kim

Hansen

Chiriboga

D. A.

(2007). Attitudes of older Korean Americans toward mental health services. Journal of the American Geriatrics Society, 55, 616-620.

23.

King

(1999). Denver American Indian mental health needs survey. American Indian and Alaska Native Mental Health Research, 8(3), 1-12.

24.

Komiya

Good

G. E.

Sherrod

N. B.

(2000). Emotional openness as a predictor of college students’ attitudes toward seeking psychological help. Journal of Counseling Psychology, 47, 138-143.

25.

Kulis

Napoli

Marsiglia

F. F.

(2002). Ethnic pride, biculturalism, and drug use norms of urban American Indian adolescents. Social Work Research, 26, 101-112.

26.

Lee

K. H.

Woo

(2013). Stressors, social support, religious practice, and general well-being among Korean adult immigrants. Journal of Evidence-Based Social Work, 10, 421-434.

27.

Lindamer

L. A.

Liu

Sommerfeld

D. H.

Folsom

D. P.

Hawthorne

Garcia

. . .Jeste

D. V.

(2012). Predisposing, enabling, and need factors associated with high service use in a public mental health system. Administration and Policy in Mental Health and Mental Health Services Research, 39, 200-209.

28.

Manson

S. M.

(2000). Mental health services for American Indians and Alaska Natives: Need, use, and barriers to effective care. Canadian Journal of Psychiatry, 45, 617-626.

29.

Mertler

C. A.

Vannatta

R. A.

(2002). Advanced and multivariate statistical methods. Los Angeles, CA: Pyrczak.

30.

Novins

D. K.

Beals

Shore

J. H.

Manson

S. M.

(1996). Substance abuse treatment of American Indian adolescents: Comorbid symptomatology, gender differences, and treatment patterns. Journal of the American Academy of Child & Adolescent Psychiatry, 35, 1593-1601.

31.

Price

B. K.

McNeill

B. W.

(1992). Cultural commitment and attitudes toward seeking counseling services in American Indian college students. Professional Psychology: Research and Practice, 23, 376-381.

32.

Robin

R. W.

Chester

Rasmussen

J. K.

Jaranson

J. M.

Goldman

(1997). Factors influencing utilization of mental health and substance abuse services by American Indian men and women. Psychiatric Services, 48, 826-832.

33.

Sarche

Spicer

(2008). Poverty and health disparities for American Indian and Alaska Native children. Annals of the New York Academy of Sciences, 1136(1), 126-136.

34.

Singh

Misra

(2009). Loneliness, depression and sociability in old age. Industrial Psychiatry Journal, 18, 51-55.

35.

Snowden

L. R.

Yamada

(2005). Cultural differences in access to care. Annual Review of Clinical Psychology, 1, 143-166.

36.

Stockdale

S. E.

Tang

Zhang

Belin

T. R.

Wells

K. B.

(2007). The effects of health sector market factors and vulnerable group membership on access to alcohol, drug, and mental health care. Health Services Research, 42(3, Pt. 1), 1020-1041.

37.

Surood

Lai

D. W.

(2010). Impact of culture on use of western health services by older South Asian Canadians. Canadian Journal of Public Health, 101, 176-180.

38.

Tijhuis

Peters

Foets

(1990). An orientation toward help-seeking for emotional problems. Social Science & Medicine, 31, 989-995.

39.

University of Maryland Center for School Mental Health. (2011). Promoting wellness in American Indian youth. Baltimore, MD: University of Maryland School of Medicine.

40.

U.S. Census. (2010). The American Indian and Alaska Native population: 2010. Retrieved from http://www.census.gov/prod/cen2010/briefs/c2010br-10.pdf

41.

U.S. Census Bureau. (2011). 2010 Census urban and rural classification and urban area criteria. Retrieved from http://www.census.gov/geo/www/ua/2010urbanruralclass.html

42.

Walls

M.L.

Johnson

K.D.

Whitbeck

L.B.

Hoyt

D.R.

(2006). Mental health and substance abuse services preferences among American Indian people of the northern Midwest. Community Mental Health Journal, 42 (6), 521-535.

43.

West

Williams

Suzukovich

Strangeman

Novins

(2012). A mental health needs assessment of urban American Indian youth and families. American Journal of Community Psychology, 49, 441-453.

44.

Westermeyer

Canive

Thuras

Chesness

Thompson

(2002). Perceived barriers to VA mental health care among Upper Midwest American Indian veterans: Description and associations. Medical Care, 40(Suppl. 1), 162-171.

45.

Zimet

G. D.

Dahlem

N. W.

Zimet

S. G.

Farley

G. K.

(1988). The Multidimensional Scale of Perceived Social Support. Journal of Personality Assessment, 52, 30-41.

46.

Zuckerman

Haley

Roubideaux

Lillie-Blanton

(2004). Health service access, use, and insurance coverage among American Indians/Alaska Natives and whites: What role does the Indian health service play? American Journal of Public Health, 94, 53-59.