Integrating Clients' Religion/Spirituality Into Practice: A Comparison Between Psychologists, Counselors, Marriage and Family Therapists, and Clinical Social Workers in Colorado

ABSTRACT

Objective

This study sought to examine the attitudes and behaviors about integrating client religion/spirituality (RS) into clinical practice for four mental health professions in Colorado.

Method

A cross‐sectional design was used consisting of the Religious/Spiritually Integrated Practice Assessment Scale and background questions to survey 619 licensed professional counselors, clinical social workers, psychologists, and marriage and family therapists living in Colorado. About 77% of respondents were cisgendered white women who reported an average age of 50, and over 75% had a personal religious affiliation.

Results

Profession was significantly associated with clinicians' attitudes about integration of client RS (partial η² = 0.039), with psychologists showing significantly lower positive attitudes than the other professions and a significantly lower overall RS integration score than counselors and marriage and family therapists. Compared with counselors, psychologists also reported a lower frequency of implementing several specific RS integration behaviors. Our results did not replicate the findings of between profession differences from a prior study in Texas.

Conclusions

The sampled psychologists in Colorado may have less favorable attitudes about integrating client RS into clinical practice compared with other mental health professionals. New education initiatives, especially those sharing the positive and growing evidence‐base for R/S‐integrative therapies and interventions, may be useful in promoting more positive attitudes amongst these Colorado psychologists if their negative attitudes can be overcome. Reviewing the results of this study in light of previous research conducted in Texas (Oxhandler and Parrish 2018) suggests geographic heterogeneity in integrating RS into practice.

Incorporating clients' RS into clinical practice improves retention and treatment outcomes (Captari et al. 2018, 2022; Swift et al. 2022) and RS‐based therapy is moderately more effective than regular treatment for individuals with strong RS affiliation (Bouwhuis‐Van Keulen et al. 2023). Numerous RS psychotherapeutic interventions exist backed by rigorous research evidence (Captari et al. 2018, 2022; Hook et al. 2010; Smith et al. 2007). Unfortunately, there can be significant hesitancy amongst mental health professionals to assess client RS, address client RS in session, or apply one of the many evidence‐based RS‐integrative interventions in existence (Oxhandler et al. 2015; Post et al. 2013; Sheridan et al. 1992; Willman et al. 2023). This is despite an ethical obligation to consider client RS in treatment across mental health professions (American Association for Marriage and Family Therapy 2015; American Counseling Association 2014; American Psychological Association 2017; National Association of Social Workers 2021). It has also been found that there are differences between mental health professions in the extent and patterns of integrating RS into practice in the United States (e.g., Bergin and Jensen 1990; Sheridan et al. 1992) and Canada (Willman et al. 2023).

If indeed some mental health professions tend to produce particularly hesitant clinicians when it comes to RS integration, this information would be relevant for both organizations that provide or oversee this training and prospective clients. Some professions may need to consider alterations to how they train and encourage their clinicians to approach RS with clients. Additionally, some potential clients may wish to know about reliable discrepancies between professions to inform their selection of a mental health professional who would be more likely to value and incorporate their RS into sessions.

To date, most of the research identifying differences between professions in how they approach RS relied on non‐standardized measures without a demonstrated history of reliability and validity—choosing often to rely on single‐item measures or study‐specific questionnaires. This provides some uncertainty about the veracity of their results. In addition, national surveys (e.g., Bergin and Jensen 1990) assume there are no notable state‐by‐state differences. Oxhandler and Parrish (2018) addressed these limitations by utilizing a standardized measure (the Religious/Spiritually Integrated Practice Assessment Scale) to investigate psychologists, counselors, marriage and family therapists (MFTs), clinical social workers (CSWs), and advanced practice nurses (APNs) specifically in Texas. After controlling for variables that differed significantly between the professions and were associated with RS (age, intrinsic religiosity, and number of prior courses in RS), they found no differences in expressed attitudes towards the importance of integrating RS into practice. However, they found significant inter‐profession differences with respect to self‐efficacy, perceived feasibility, and self‐reported behaviors with MFTs scoring highest in all four areas. Across self‐efficacy and behaviors, and in overall orientation towards integrating RS into practice, psychologists and APNs generally scored the lowest, MFTs highest, and counselors and CSWs in the middle. When specific RS‐integrative practice behaviors were assessed, the overall trend was again MFTs being most likely to report engaging in most of those behaviors. In no instance did psychologists or APNs report engaging in a specific RS‐integrative practice behavior more than MFTs, counselors, or CSWs. Oxhandler and Parrish further assessed the personal religiosity of these professionals and found significant differences between the professions favoring MFTs over all other professions with APNs having lower religiosity than all other professions except psychologists.

1. Purpose of This Study and Research Questions (RQs)

Oxhandler and Parrish (2018) compared the integration of clients' RS into clinical practice between different types of mental health professionals in Texas. Even if we assume that Oxhandler and Parrish's (2018) results are representative of mental health professionals in Texas, it is still uncertain if their results generalize to other U.S. states or are state‐specific. Our study responds to these concerns and adds to the body of literature on potential inter‐profession and inter‐state differences in client RS integration. We conducted a conceptual replication of Oxhandler and Parrish's (2018) study with four types of mental health professionals (psychologists, counselors, MFTs, and CSWs¹) in Colorado. The choice of a state other than Texas allowed us an important opportunity to examine potential variability at the regional (i.e., state) level. Similar results would be consistent with a national trend in RS‐integration across profession. Divergent results would suggest geographical variability at the state level. Our research questions mirror those from Oxhandler and Parrish (2018) except for omitting APNs: How do psychologists, counselors, MFTs, and CSWs compare in self‐reported variables related to integrating RS in psychotherapy? We specifically asked this question regarding six areas: personal religiosity (RQ1), attitudes (RQ2), self‐efficacy (RQ3), feasibility (RQ4), behaviors (RQ5), and overall orientation of integrating RS into clinical practice (RQ6). In addition, we asked how our results compare to Oxhandler and Parrish's results in Texas (RQ7).

2. Method

2.1. Sampling Procedures and Participant Recruitment

This study was approved by the institutional research ethics boards of the University of British Columbia and the University of Northern Colorado. The research team obtained active license lists for the four mental health professions from the Colorado Department of Regulatory Agencies. Practitioners licensed in Colorado but not living or practicing in Colorado (n = 3797) or those with dual licenses between two of the four professions (n = 268) were excluded from the study. This was to ensure a participant pool most representative of those providing therapy in Colorado, the intended population under consideration. We excluded an additional 67 names as invalid for other reasons (e.g., an expired license, the practitioner being deceased). This resulted in single‐license population sizes of 2431 psychologists; 6503 counselors; 5213 CSWs, and 738 MFTs. We located contact information for 12,880 of them (86.5%; 2184 psychologists [89.8%]; 6,193 counselors [95.2%]; 3912 CSWs [75.0%]; and 591 MFTs [80.1%]).

Recruitment lasted 5 months. Professionals were invited to participate in the online survey by either email, website contact, LinkedIn profile, or phone. Study information was also sent to state‐specific professional listservs. Participants were also invited to enter a random draw for a $100 Starbucks gift card. After initial contact, three follow up requests were sent 2 weeks apart for email contacts, and one follow up request for phone and website contacts. Qualtrics was the online survey platform used.

2.2. Participants

In total, 727 professionals responded. We removed 37 cases with no responses, 59 cases with only completed demographic questions, and 12 invalid cases that did not meet inclusion criteria (e.g., dual licenses or a different profession), leaving 619 valid responses for analysis (4.1% representation rate of profession population). Psychologists accounted for 131 responses (4.1% population representation rate), counselors 340 responses (4.2% representation rate), CSWs 114 (1.7% representation rate), and MFTs 34 responses (3.0% representation rate). Table 1 shows sample variables that were parallel with those of Oxhandler and Parrish's (2018) sample. Beyond what is shown in this table, our final set of participants was predominantly heterosexual at 86% (5.8% bisexual, 3.9% lesbian/gay, 3.6% queer people, and 1.0% other sexual minorities) and White/European at 82% (12% multi‐ethnicity; 3.2% other ethnicity; 2.9% Latinx/Central/South American). About 78.0% reported having a personal religious affiliation (22% no religious affiliation and 6% Atheist). Of those who reported a religion, 41% were Christian (Other 10.7%; Agnostic 7.8%; Buddhist 6.6%; Jewish 5.7%). For highest academic credential, 26% had a doctoral degree, 72% a master's degree, and 2% a bachelor's degree or diploma. Most participants had no prior RS continuing education (75.6%).

Table 1.

Sample Characteristics and Background Variables Compared with Oxhandler and Parrish (2018).

Characteristic or variable	Present Study			Oxhandler and Parrish (2018) (n = 462)
	n	M	SD	M	SD
Age	618	50.17	13.31	52.37	12.86
Years of practice experience	619	17.01	11.34	17.57	10.99
Intrinsic religiosity	597	10.55	4.10	11.75	3.41
	n	%		n	%
Gender
Female	465	77.0%		332	71.9%
Male	130	21.5%		130	28.1%
Genderqueer	9	1.5%		NR
Professional license
Counselors	340	54.9%		122	26.4%
CSWs	114	18.4%		142	30.7%
Psychologists	131	21.2%		100	21.6%
MFTs	34	5.5%		98	21.2%
Has taken courses on RS	190	30.9%		84	18.2%
Has RS agency affiliation	51	8.3%		47	10.2%

Note: CSW = clinical social workers; MFT = marriage and family therapists; RS = religion or spirituality; NR = not reported.

2.3. Measures

2.3.1. Participant Characteristics Questionnaire

A questionnaire was created that asked about personal and professional characteristics. It included the five demographic variables measured by Oxhandler and Parrish (2018).

2.3.2. Religious/Spiritually Integrated Practice Assessment Scale

The Religious/Spiritually Integrated Practice Assessment Scale (RISPAS; Oxhandler and Parrish 2016) is a 40‐item standardized measure of overall orientation towards integrating RS into clinical practice. It contains four subscales corresponding to attitudes, self‐efficacy, feasibility, and behaviors. The RSIPAS uses a Likert scale of 5 from “strongly disagree”/“never” to “strongly agree”/“very often”, with higher scores meaning greater integration of RS in clinical practice. It include questions like “I am comfortable discussing my clients' religious/spiritual struggles” and “Integrating clients' religious/spiritual beliefs in treatment helps clients meet their goals.” Oxhandler and Parrish (2018) found an overall Cronbach's α = 0.95 with subscale values ranging from 0.88 to 0.91. The internal consistency for the RSIPAS total score in the present study was α = 0.94 and for the four subscales it was Self Efficacy α = 0.89, Attitudes α = 0.85, Feasibility α = 0.84, and Behavior α = 0.85.

2.3.3. Duke University Religion Index

The Duke University Religion Index (DUREL; Koenig and Büssing 2010) is a five‐item standardized measure of religious involvement. It has three subscales: organized religious activity, non‐organized religious activity, and intrinsic religiosity (IR). Only the IR subscale, consisting of three items, was used in this study following Oxhandler and Parrish (2018; α = 0.95). The IR subscale uses a 5‐point Likert scale from “definitely not true” to “definitely true of me,” with larger scores indicating stronger intrinsic religiosity. An example item is “In my life, I experience the presence of the Divine (i.e. God).” In past research, the IR subscale has shown good convergent validity, test‐retest reliability and internal consistency (Koenig and Büssing 2010; Lace and Handal 2018). The coefficient alpha for the IR subscale in this study was α = 0.90.

2.4. Data Analysis

Descriptive and inferential statistics were run using SPSS (versions 28 and 29). Cohen's h values were calculated in Microsoft Excel. Missing data was in the acceptable range (Tabachnick and Fidell 2013) with only 1.5–2.9% of data missing per item. The data appeared to be missing completely at random (MCAR: χ² = 163.68, DF = 223, p = 0.999). Therefore, complete case analysis (i.e., list‐wise deletion) was appropriate (Parent 2013).

We tested 13 demographic variables and the DUREL IR subscale for interprofessional differences using chi‐square analyses, Fisher's exact test, or ANOVA depending on frequencies and variable type. Given the large number of potential exploratory covariates based on how many demographic and background questions were asked, to reduce Type 1 error due to our large sample size and improve statistical degrees of freedom, two criteria had to be met to be considered an exploratory covariate variable for analysis in this study. First, there had to be statistically significant group differences found in a majority (i.e., > 2) of the four RSIPAS subscales. Second, the effect size of the group difference had to have at least a small effect size (r ≥ 0.1 or η ²  ≥ 0.02) based on Cohen 1992] criteria.

Standard one‐way multivariate analysis of covariance (MANCOVA) assumptions were assessed (Shapiro‐Wilk test of normality, homogeneity of variance and homogeneity of regression slopes, correlation between covariates and dependent variables). Only the test of normality was violated. However; MANCOVA is robust to violations of multivariate normality at sample sizes of this magnitude (Olejnik and Algina 1984). Therefore, mirroring Oxhandler and Parrish (2018), a MANCOVA was run to compare the four professions' responses to each of the RSIPAS's subscales and control for covariates. To gauge the impact of including covariates, we repeated this analysis with a multivariate analysis of variance (MANOVA) that excluded covariates. A one‐way analysis of variance (ANOVA) was then used to examine the relationship between profession and total RSIPAS score. To investigate any specific differences in behavior, chi‐square tests were used to evaluate associations between the professions and responses on the 9‐item behavior subscale with responses to the behavioral questions collapsed to never/rarely, some of the time, and often/very often (cf. Oxhandler and Parrish 2018).

2.5. Results

A comparison of the professions across the four observed covariates (years in practice, RS coursework, intrinsic religiosity, religious affiliation) that met our inclusion criteria is presented in Tables 2 and 3. The correlations of RSIPAS total score and subscale scores with these observed covariates is presented in Table 4. Stringent Bonferroni‐corrected post hoc comparisons for pairwise differences revealed differences among the professions for years in practice, intrinsic religiosity, prior RS coursework and religious affiliation based on a study‐wise alpha of 0.05. Psychologists reported more years of practice experience than members of all other professions (counselors: p < 0.001, Cohen's d = 0.69; CSWs: p = 0.004; Cohen's d = 0.32; MFTs: p = 0.005; Cohen's d = 0.47). Counselors averaged fewer years in practice than CSWs (p = 0.004; Cohen's d = 0.33). Counselors also reported greater intrinsic religiosity than CSWs (p = 0.002; Cohen's d = 0.35) and psychologists (p < 0.001; Cohen's d = 0.38). Prior coursework on integrating RS was relatively rare for CSWs and psychologists and significantly more common (but still not the norm) for counselors (p < 0.001 for both comparisons; Cohen's h = 0.50 for CSWs and 0.46 for psychologists). The only significant difference in religious affiliation was that Judaism was more common among psychologists than counselors (p = 0.002; Cohen's h = 0.29; small effect).

Table 2.

Observed Scale Covariates across Professions.

Variable	Counselors		CSWs		Psychologists		MFTs		N	F	η2
	M	SD	M	SD	M	SD	M	SD
Years of experience	14.72	9.61	18.11	11.98	22.21	13.45	16.26	9.57	619	15.20***	0.069
DUREL intrinsic religiosity	11.11	3.87	9.72	4.37	9.58	4.41	11.52	3.24	597	6.67***	0.033

Note: CSW = clinical social workers; MFT = marriage and family therapists. Welch's ANOVA was used because both variables failed the assumption of homogeneity of variances. Significance level conclusions were the same for both classic and Welch's ANOVA.

^*** p < 0.001.

Table 3.

Observed Categorical Covariates across Professions.

Demographic	Counselors		CSWs		Psychologists		MFTs		χ2	p	V
	N	%	n	%	N	%	n	%
Religious affiliation									29.55	0.042*	0.126
Unaffiliated	74	21.8	25	21.9	29	22.1	9	26.5
Atheist	15	4.4	11	9.6	11	8.4	0	0.0
Christian	143	42.1	52	45.6	43	32.8	17	50.0
Jewish	12	3.5	7	6.1	14	10.7	2	5.9
Agnostic	24	7.1	8	7.0	13	9.9	3	8.8
Buddhist	28	8.2	5	4.4	7	5.3	1	2.9
Other	44	12.9	6	5.3	14	10.7	2	5.9
Prior courses on RS									30.68	< 0.001***	0.223
Yes	135	39.7	20	17.5	25	19.1	12	35.3

Note: RS = religion or spirituality; LPC = licensed professional counselors; CSW = clinical social workers; MFT = marriage and family therapists; V represents Cramer's V; CEUs = continuing education units.

^* p < 0.05

^** p < 0.01

^*** p < 0.001.

Table 4.

Correlations of RSIPAS with Observed Covariates.

Variable	RSIPAS subscale				RSIPAS total
	Self‐Efficacy	Attitudes	Feasibility	Frequency
Years in practice	0.24***	0.05	0.22***	0.13**	0.18***
RS coursework	0.29***	0.21***	0.26***	0.33***	0.33***
DUREL intrinsic religiosity	0.39***	0.39***	0.33***	0.43***	0.46***
Religious affiliationa	0.29***	0.26***	0.17**	0.32***	0.31***

Note: RSIPAS = Religious/Spiritually Integrated Practice Assessment Scale.

^a For this nominal variable, eta correlation coefficients are shown in place of Pearson's r.

^** p < 0.01

^*** p < 0.001.

Box's M test yielded a result consistent with equality of covariance matrices; F(30, 52780.837) = 1.716, p = 0.009. The RSIPAS subscales were significantly associated with profession after accounting for the covariates, F(12, 1526.89) = 2.982, p < 0.001, Wilks' Λ = 0.941, partial η² = 0.020. The MANOVA analysis significantly predicted scores on the RSIPAS subscales, F(12, 1561.28) = 3.98, p < 0.001, Wilks' Λ = 0.923, partial η² = 0.026. Thus, profession differences are present both with and without accounting for covariates. Means and results of RSIPAS subscale‐specific comparisons from the MANCOVA are shown in Table 5. After accounting for the covariates, profession was only a significant predictor for attitudes (p < 0.001; partial η² = 0.039, small effect size). Bonferroni‐corrected pairwise comparisons indicated that psychologists had lower attitude scores compared to the other professions (p < 0.001 with CSWs, p = 0.002 with counselors, and p = 0.005 with LMFTs), but the attitude scores of the professions other than psychologists did not differ significantly from each other.

Table 5.

Mean Differences Between Disciplines on RSIPAS Subscale Scores After Accounting for Covariates.

RSIPAS subscale	Counselors (n = 323)		CSWs (n = 113)		Psychologists (n = 125)		MFTs (n = 32)		F	p	Partial η2
	M	SD	M	SD	M	SD	M	SD
Self‐efficacy	53.48	6.50	51.75	7.93	52.19	7.42	54.94	6.30	1.42	0.236	0.007
Attitudes	50.24	5.44	50.37	5.30	47.26	6.94	51.41	6.41	7.95	<0.001	0.039
Perceived feasibility	23.85	3.61	23.12	4.23	23.02	4.51	25.00	3.98	2.15	0.093	0.011
Self‐reported behaviors	28.88	6.30	27.27	6.89	26.78	6.68	28.75	6.14	.82	0.483	0.004

Note: RSIPAS = Religious/Spiritually Integrated Practice Assessment Scale; CSW = clinical social workers; MFT = marriage and family therapists.

A one‐way ANOVA indicated that the professional groups also differed on the RSIPAS overall test score (designating overall RS orientation) with a small effect size: F(3, 593) = 5.53, p = < 0.001, η² = 0.027. Post hoc pairwise comparisons with a Bonferroni correction showed that the only two significant differences were between psychologists (M = 149.21, SD = 21.73, n = 147) and counselors (M = 156.58, SD = 18.20, n = 326) at p < 0.001 with a small effect size (Cohen's d = 0.38) and between psychologists and MFTs (M = 160.09, SD = 20.05, n = 32) at p = 0.005 with a moderate effect size (Cohen's d = 0.51). Counselors, MFTs, and CSWs did not significantly differ from each other, and CSWs did not differ from psychologists.

Five of the RSIPAS behaviors subscale items showed a significant difference, all with small effect sizes (see results in Table 6). When compared with psychologists, counselors reported more frequently seeking out consultation about addressing clients' RS and more often used empirically supported RS interventions (p < 0.001, Cohen's h = 0.44 for never/rarely). Similarly, counselors were more likely than psychologists to report often or very often involving clients in decisions of whether to integrate their RS into clinical practice (p = 0.003, Cohen's h = 30). Counselors were reportedly more frequent than CSWs in helping their clients consider the RS‐related meaning of the client's life situation (p = 0.001, Cohen's h = 0.32 for never/rarely).

Table 6.

“Often” or “Very Often” Responses to Behavioral Scale Items, Cross‐Tabulated by Discipline.

Behavior item	Counselors (n = 328‐329)		CSWs (n = 113)		Psychologists (n = 127)		MFTs (n = 33)		χ2	p	V
	n	%	n	%	n	%	n	%
1) I seek out consultation on how to address clients’ religious/spiritual issues in treatment.	48	14.6	12	10.6	8	6.3	3	9.1	12.81	0.046	0.103
2) I read about ways to integrate clients’ religion/spirituality to guide my practice decisions.	93	28.4	23	20.4	21	16.5	5	15.2	14.84	0.022	0.111
3) I read about research evidence on religion/spirituality and its relationship to health to guide my practice decisions.	55	16.8	11	9.7	18	14.2	2	6.1	9.67	0.139	0.090
4) I involve clients in deciding whether their religious/spiritual beliefs should be integrated into their treatment.	250	76.2	76	67.3	79	62.2	28	84.8	13.71	0.033	0.107
5) I use empirically supported interventions that specifically outline how to integrate my clients’ religion/spirituality into treatment.	56	17.1	14	12.4	13	10.2	4	12.1	19.09	0.004	0.126
6) I conduct a full biopsychosocialspiritual assessment with each of my clients.	133	40.5	57	50.4	47	37.0	18	54.5	7.49	0.278	0.079
7) I link clients with religious/spiritual resources when it may potentially help them.	173	52.6	55	48.7	50	39.4	15	45.5	7.38	0.287	0.078
8) I help clients consider ways their religious/spiritual support systems may be helpful.	240	72.9	70	61.9	90	70.9	21	63.6	a	0.073	0.095
9) I help clients consider the religious/spiritual meaning and purpose of their current life situations.	188	57.1	61	54.0	69	54.3	21	63.6	12.99	0.043	0.104

Note: CSW = clinical social workers; MFT = marriage and family therapists; V represents Cramer's V.

^a Due to the number of cells with expected counts less than 5, Fisher's exact test was used. Thus, there is no chi statistic.

3. Discussion

3.1. Religiosity Across Professions

The current study further strengthens the already well‐replicated finding that psychologists tend to report being less personally religious/spiritual than other mental health professionals (e.g., Bergin and Jensen 1990; Mandelkow et al. 2022; Oxhandler and Parrish 2018; Sheridan et al. 1992; Willman et al. 2023). For the sampled mental health professionals in Colorado, MFTs followed by counselors reported the highest levels of personal religiosity with evidence that both were more religious than psychologists and CSWs.

The finding of lower religiosity amongst certain professions could result from some combination of three processes: (a) changes within individuals after joining a profession such as students becoming less religious during their training (Schwadel 2016); (b) professions such as psychology discriminating against highly religious students in the admissions process (Gartner 1986; Honeycutt and Freberg 2017; Ressler and Hodge 2006; Thyer and Myers 2009; Yancey 2011); and (c) self‐selection by prospective mental health professionals such as highly religious students avoiding psychology or the additional years required of doctoral level study, which is the entry level in psychology (e.g., Noll 1994). It could also be that religious students are opting into training programs that emphasize RS integration into professional practice. If such programs are more common for counselors and MFTs than for CSWs and psychologists, it could help explain the differences in IR among professions. Consistent with this idea, in both our study and that of Oxhandler and Parrish (2018), greater than 25% of both LPCs and MFTs reported having had prior courses on RS (which suggests many attended RS‐integrative training programs), whereas less than 20% of either CSWs or psychologists reported taking such courses.

3.2. Interprofessional Differences Related to Integrating RS into Clinical Practice

Personal religiosity was the strongest covariate predictor of all four components of integrating RS into practice with medium to nearly large effect sizes. Therefore, it should not come as a surprise that the profession with the lowest personal religiosity (psychologists) also showed the lowest amount of integrating RS into clinical practice. With respect to overall orientation towards integrating RS into clinical practice, psychologists scored significantly lower than both counselors and MFTs. Accounting for profession differences on covariates such as intrinsic religiosity, the only difference found across professions on the RSIPAS subscales pertained to attitudes. Psychologists had the least favorable attitudes towards integrating RS into clinical practice compared to counselors, MFTs, and CSWs. In terms of specific ways of incorporating RS into clinical practice, compared to psychologists, counselors more often sought consultation regarding client RS issues, incorporated evidence‐based RS interventions into session, and actively involved clients in decision‐making about RS in session. There was also evidence that, compared to CSWs, counselors more often helped clients consider the RS meaning of their current life situation.

3.3. Comparison with Oxhandler and Parrish (2018)

Oxhandler and Parrish (2018) in their study of mental health professions in Texas found three variables that differed between the professions and predicted domains of integrating RS into clinical practice: age, religiosity, and prior coursework on RS. The present study replicated religiosity and RS coursework and did not replicate age. Therefore, there is replicated evidence that personal religiosity and prior RS coursework are key pre‐existing differences between the professions that cross state lines. By extension, this implies that differences across professions with regard to the personal religiosity of practitioners and provision of RS coursework could affect how much a particular group of professionals integrates RS into clinical practice. Therefore, the lower positive attitude and overall orientation for psychologists towards integrating RS into practice might result from the collectively low personal religiosity among psychologists shaping the norms of the discipline and courses offered.

Most of the other findings of the present study appear at first glance to be in direct opposition to those found by Oxhandler and Parrish (2018) in Texas. Oxhandler and Parrish found interprofessional differences between professions with respect to self‐efficacy, feasibility, behaviors, and overall orientation; and no differences in attitudes towards integrating RS into clinical practice. The present study, conducted with Colorado professionals, found differences in attitudes and overall orientation but not in the other three domains. Even in overall RS orientation, there were slight differences. Oxhandler and Parrish found MFTs stronger in overall orientation than social workers and psychologist but not counselors and no difference between social workers, counselors, and psychologists. The present study found both MFTs and counselors stronger in overall orientation than psychologists but no differences between counselors, CSWs, and MFTs. Thus, for overall orientation, the only replicated finding was that MFTs were more oriented towards integrating RS into clinical practice than psychologists.

In Oxhandler and Parrish (2018), MFTs showed higher scores for overall orientation, self‐efficacy, feasibility, and self‐reported behaviors than other disciplines. Although we failed to replicate this pattern, we attribute this to a higher subgroup sample size and thus statistical power for MFTs in Oxhandler and Parrish. The pattern of means we observed in the present study was consistent with MFTs having comparable or greater scores on all scales/subscales compared to all other professions. Other than this, the main difference between the two studies with respect to the RSIPAS is that we found psychologists were modestly lower on the attitudes scale than the other professions, while Oxhandler and Parrish did not.

Two sources of explanation for this difference in attitudes include differences in region and sample characteristics. First, it is possible the divergent results could represent distinct between‐profession differences in integrating RS into clinical practice in Texas versus Colorado. Potentially any difference between the two states could influence this. Among the most plausible is overall religiosity; Texas seems to be more religious than Colorado (Pew Research Center 2014), which is consistent with the intrinsic religiosity found in the two study samples. Based on Namaste Theory (RS individuals are more likely to appreciate RS in others; Oxhandler 2017), we should expect more RS‐integration in psychologists in Texas who are presumably more religious. As shown in Table 1, the sample demographics of the two studies were quite comparable except our participants were a couple of years younger on average and more frequently female than those of Oxhandler and Parrish (2018), although it is not clear why this would cause a change in results.

Nevertheless, there are several reasons why we believe our results are more trustworthy and robust compared to Oxhandler and Parrish's. First, Oxhandler and Parrish's final sample size was only 536 participants, and this represented only 1.4% of the population of the specified mental health professions. In contrast, our study used 619 participants and this represented nearly 3 times more of the state population (4.2%). Second, Oxhandler and Parrish had a more limited sampling frame, which reduced the generalizability of their data. They attempted to contact only 8.4% (n = 3500) of all licensed mental health professionals. In contrast, our sampling frame included all single‐licensed professionals in the stated professions (n = 14,885) and we were able to locate contact information and contact about 87% of them (n = 12,880). Third, Oxhandler and Parrish's study would have included dual‐licensed professionals clouding the distinctness between the professional due to overlapping professionals and violating the statistical independence needed for their statistical analyses. In contrast, our study identified 268 professionals who held more than one license (about 1.8% of all professionals), who were excluded from between‐profession analyses.

3.4. Limitations of the Study and Constraints to Generalizability

Although this present study's sample size and population representation rate notably exceeded that of Oxhandler and Parrish (2018), and we attempted to contact all relevant professionals in Colorado, we still only sampled about 4.1% of them. In addition, the present study (like Oxhandler and Parrish 2018) did not randomly sample from the population or administer a survey to the entire population of mental health professionals in the state. Therefore, participant self‐selection bias cannot be ruled out for the particular results obtained.

3.5. Future Research

This study should be replicated in Colorado to assess the robustness of its findings, especially by oversampling MFTs relative to their proportion in the population to obtain more statistical power to detect MFT differences. Working directly with the state licensure body or professional associations could help increase sample size. It is also important to complete this study again in a different state to further assess the extent of generalizability of the present study results and those of Oxhandler and Parrish (2018). Further, dual license holders remain uninvestigated with respect to integrating RS into clinical practice. It could be that holding dual licenses could result in a greater likelihood of receiving RS coursework, which should translate into enhanced willingness and ability to incorporate RS into clinical practice.

3.6. Implications of the Results

Previous coursework in RS and personal religiosity were both found to be reliable predictors of components of integrating RS into clinical practice in the present study as well as in Oxhandler and Parrish (2018). Therefore, the simplest solution to increasing practitioner self‐efficacy, self‐reported behaviors, and perhaps even perceived feasibility and attitudes could be increasing the amount of coursework and clinical supervision related to integrating RS into professional practice. In the present study, prior RS coursework was not common across any of the professionals, but was especially rare for social workers and psychologists. Although it is possible to infuse training in any of the many and growing evidence‐based RS interventions and therapies across various courses (Captari et al. 2018, 2022; Hook et al. 2010; Smith et al. 2007), another possibility would be to offer a dedicated course related to developing RS clinical competence. However, the lower overall positive attitudes that psychologists have with respect to integrating RS into practice could serve as a barrier towards enacting these suggestions. Based on Namaste Theory (RS individuals are more likely to appreciate RS in others; Oxhandler 2017), another strategy towards increasing the integration of RS into clinical practice would be to increase the number of RS practitioners (Oxhandler 2017). However, it may be necessary to first address negative attitudes and prejudice in admissions against highly religious individuals (Yancey 2011; Honeycutt and Freberg 2017). Assuming the veracity of the present study's results, an argument can also be made for informed consent. At least in Colorado, prospective clients should be aware that they are most likely to find a professional who is willing and able to bring RS into sessions if they seek services from a counselor (and perhaps MFT).

4. Conclusion

In Colorado, it appears that psychologists have the least favorable attitudes towards integrating RS into clinical practice and also the lowest overall orientation towards integrating RS into clinical practice. It may be necessary to first address neutral or ambivalent attitudes about the integration of RS into practice, particularly amongst psychologists (Willman et al. 2023) before other initiatives, such as offering more RS‐oriented courses or recruiting more RS individuals into psychology programs, could be effective. Comparing our results with previous research, we have no reason to believe there are consistent national trends in interprofessional differences related to integrating RS into clinical practice but we do have reason to believe that psychologists and CSWs are overall less RS.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.