Abstract
Barriers to ethical and effective health care in rural communities have been well-documented; however, less is known about strategies rural providers use to overcome such barriers. is study compared adaptations by rural and urban health care providers. Physical and behavioral health care providers were randomly selected from licensure lists for eight groups to complete a survey; 1,546 (52%) responded. Replies indicated that health care providers from small rural and rural communities were more likely to integrate community resources, individualize treatment recommendations, safeguard client confidentiality, seek out additional expertise, and adjust treatment styles than were providers from small urban and urban communities. Behavioral health care providers were more likely than physical health care providers to integrate community resources, individualize treatment recommendations, safeguard client confidentiality, and adjust their treatment styles; physical health care providers were more likely than behavioral health care providers to make attempts or have options to seek out additional expertise.
Keywords: Adaptations, rural, ethics, health care barriers
Recent research has shown that rural residents and health care providers face a number of barriers associated with the provision of ethical and effective health care service,1 related to rural resource limitations, challenges to confidentiality, overlapping roles for patients and providers, travel and geographic complications, service access restrictions, and limitations in training and education. Some barriers are clearly related to the special circumstances of rural life, some are unique to the nature of rural residents, and some are closely tied to the experiences and backgrounds of rural health care providers.
Combined, these challenges represent and lead to health and health care disparities2,3 that perpetuate lower health-related quality of life for rural residents,4 especially rural residents in racial/ethnic minority groups.5 For example, as a result of rural health care barriers, rural areas struggle with inadequate disease prevention, delayed detection of illness, mis-diagnosis and late diagnosis, inadequate referral, and ineffective treatment.6–9 Not surprisingly, repeated calls have been made for systemic interventions and training interventions that will address barriers to care in rural areas. Adaptations have been suggested ranging from increasing community awareness of prevention and early detection, to integrating community resources, implementing non-traditional treatment approaches, initiating measures to minimize confidentiality concerns, and finding ongoing support resources for providers.10–13 Included as an appendix is a listing of some of the more common recommendations; this list reflects the major themes in the literature,14,15 that is, system integration and flexibility, community and professional outreach, and cultural sensitivity and awareness.
Such changes, however, have been slow to occur, in part due to the rigidity of higher educational systems (such as medical and graduate schools), the inflexibility of public policy, and the intransigence of systems in general.2,3,14,15 To cope with the challenges they perceive in their daily practice, rural providers have learned to make adaptations to the ways they were taught to deliver services in an attempt to overcome some of the systemic, community, patient, and educational challenges that impede optimal health care. These modifications to ‘business as usual’ are beginning to be documented in the anecdotal literature, but have not been explored systematically. Some of the documented adaptations by individuals include, but almost certainly are not limited to, the following practices:3,16,17
Keeping a private log of information about a client that is not part of the offcial clinical records
seeking support from local elders
involving the patient’s family member(s) in health care
changing personal traits to adapt to patient preferences
modifying clinic procedures
expanding regular office hours
seeking relevant continuing education
Purpose of current study
Knowing what adaptations rural care providers have made successfully would provide fertile ground for policymakers in their efforts to plan system-wide changes. However, no study in the peer-reviewed literature has formally explored the types of adaptations rural versus urban care providers have made on a regular basis in the interests of their patients’ health and welfare. To address this void in the literature, the current study sought to explore adaptations made by health care providers in their daily practice in Alaska and New Mexico with an array of population densities, ranging from small, inaccessible rural communities to relatively large urban centers. This diversity in density allowed for comparisons of adaptations used by health care providers in a variety of rural and urban communities. To accomplish the goals of the study, a survey was developed for and sent to over 3,500 health care providers, including general practice physicians, psychiatrists, physician assistants, nurse practitioners, registered nurses, psychologists, mental health counselors, and social workers. Examining current adaptations used by these health care providers will allow for well-informed design of interventions that promote awareness and provide information regarding available and utilized methods to meet the needs of consumers, differentiating methods used by community size and type of care provider.
Methods
Participants
Potential participants in Alaska and New Mexico were randomly selected from the licensure lists for 8 health care provider groups: primary care physicians, psychiatrists, physician assistants, nurse practitioners, registered nurses, psychologists, social workers, and mental health counselors. For selection purposes, all licensees in the 8 provider groups were categorized as urban or rural based on their addresses (provided by the licensing boards). Next, for each of the 8 provider groups, 125 rural and 125 urban participants were selected from each state. However, as some groups did not have 125 individuals (e.g., rural psychiatrists), the entire group was sampled and the parallel group (e.g., urban psychiatrists) was sampled to the degree necessary to total 250. Even with sampling all potential participants, some groups did not have enough providers to total 250; thus, the final identified sample was 3,695.
Of 3,695 potential participants contacted, 1,546 (723 from Alaska and 823 from New Mexico) provided useable data. useable overall response rate for the survey was 52.2%, with 50.6% in Alaska and 53.7% in New Mexico. The rural response rate was 49.8% (53.3% in Alaska and 47.3% in New Mexico), for a total of 662 rural respondents; the urban response rate was 54.1% (48.7% in Alaska and 60.4% in New Mexico), for a total of 896 urban respondents. Table 1 provides participants’ demographic information, organized by state.
Table 1.
PARTICIPANT DEMOGRAPHICS BY STATE
| Demographic variable | Alaska
n (%) |
New Mexico
n (%) |
|---|---|---|
| Total sample size | 723 (46.8) | 823 (53.2) |
| Professional affliation | ||
| Physician | 77 (10.7) | 73 (8.9) |
| Psychiatrist | 28 (3.9) | 99 (6.4) |
| Physician assistant | 105 (14.5) | 124 (15.1) |
| Nurse practitioner | 120 (16.6) | 130 (15.8) |
| Registered nurse | 113 (15.6) | 108 (13.1) |
| Psychologist | 62 (8.6) | 101 (12.3) |
| Social worker | 104 (14.4) | 105 (12.8) |
| Mental health counselor | 114 (15.8) | 111 (13.5) |
| Gender | ||
| Men | 225 (31.1) | 279 (33.9) |
| Women | 498 (68.9) | 544 (66.1) |
| Hispanic descent | ||
| Yes | 19 (2.6) | 134 (16.4) |
| No | 696 (96.3) | 686 (83.4) |
| Ethnicity | ||
| African American | 12 (1.7) | 18 (1.2) |
| Alaska Native | 13 (1.8) | 0 (0) |
| American Indian | 30 (4.1) | 32 (3.9) |
| Asian American | 6 (0.8) | 12 (1.5) |
| White | 621 (40.2) | 625 (75.9) |
| Other | 12 (1.7) | 14 (1.7) |
| No response | 12 (1.7) | 5 (0.6) |
| Marital status | ||
| Married | 536 (74.5) | 618 (74.9) |
| Single | 83 (11.5) | 106 (12.9) |
| Divorced/widowed | 100 (13.9) | 98 (11.9) |
| No response | 4 (.6) | 1 (0.1) |
| Age in years | ||
| Mean (SD) | 48.8 (9.17) | 49.0 (9.8) |
| Community of residence | ||
| Urban | 420 (58.1) | 477 (58.0) |
| Rural | 303 (41.9) | 346 (42.0) |
| Current community | ||
| Years lived in Mean (SD) | 13.1 (10.8) | 15.5 (12.2) |
| Providing care in | 622 (86.0) | 625 (73.9) |
| Lived in rural community
(1 year or more) | ||
| Yes | 630 (88.2) | 665 (82.0) |
| No | 84 (11.8) | 146 (18.0) |
| How long in years Mean (SD) | 20.7 (14.6) | 20.3 (13.9) |
| Worked in rural community
(1 year or more) | ||
| Yes | 585 (81.5) | 600 (73.8) |
| No | 133 (18.5) | 213 (26.2) |
| How long in years Mean (SD) | 10.8 (8.6) | 10.5 (8.9) |
Instrument
Following procedures recommended by Dillman,18 participants were asked to complete a survey (described below) that asked them to describe their experiences in providing physical or mental health care to their clients/patients. Survey procedures began with a letter informing the potential participants that they had been selected to receive a survey and to expect it to arrive by mail. Two weeks later, a packet consisting of an informed consent form, survey, non-respondent form, and payment form was mailed to all potential participants. Participants were informed that they would be paid $50 for returning a completed survey and that the source of funding for this project was the National Institute on Drug Abuse. Four weeks later, a reminder card was sent out to potential participants who had not returned a survey or a non-respondent card. Another complete packet was sent two weeks later to potential participants who had not returned a completed survey or non-respondent card, followed by another reminder card two weeks later.
The survey was developed based on extensive qualitative work conducted over a 2-year period. is qualitative work included 82 key informant interviews and 25 focus groups (with a total of 163 participants) conducted with health care professionals, patient advocates, and health care administrators in Alaska and New Mexico. These qualitative data were analyzed to identify convergent and divergent themes that represent the constellation of issues surrounding rural practice in general, issues with regard to rural practice vis-à-vis ethics and stigma, and issues of community consent and rural health disparities. Combining this information with existing literature, we constructed and piloted items to develop a 21-page survey. This survey included 336 items separated into the following 10 sections: ethical challenges, perceptions of illness stigma, training and resource needs, experiences in providing health care, barriers faced in providing care, adaptations to barriers, treatment issues related to providing care to minority groups, resource needs, training needs, and provider practice characteristics. Individual items were rated on an 11-point rating scale, ranging from 0 (not at all or never) to 10 (very much or always). The survey required approximately 60 to 90 minutes to complete. The survey has been the basis for several manuscripts, each drawing on separate survey sections (e.g., barriers,1 minority health care disparities,5 training needs,16 caseloads19).
The current study involved the analysis of responses to the items included in the section of the survey that addresses informal practices utilized by health care professionals to adapt to barriers to care. For these items, participants were asked to rate on an 11-point scale that ranged from 0 (Never) to 10 (Always) how often they used each of 37 different adaptations to barriers encountered in providing ethical and effective health care. Examples of such items include spending extra time with clients, making referrals, establishing a referral network, and meeting clients after hours. All 37 items are listed in Table 2 and were derived from the extant conceptual literature and extensive interviews and focus groups with providers in both states.
Table 2.
PRINCIPAL COMPONENT ANALYSIS FOR PROVIDER ADAPTATIONS
| Scales and items | Factor loading |
|---|---|
| Integration of community resources, α = 5.88 | |
| Consulted community leaders or elders to learn about ethnic/cultural issues in providing care | .73 |
| Consulted community leaders or elders to increase your ability to provide effective health care | .72 |
| Participated in community events to better understand values and customs of your clients/patients | .70 |
| Worked with sources in your community (e.g. churches, schools, law enforcement) to increase support for you clients/patients | .66 |
| Spent time learning more about community resources | .58 |
| Provided community level health education | .54 |
| Talked with clients/patients about the challenges of assuring confidentiality in their community | .54 |
| Encouraged people in your community to travel to other communities for care to safeguard their confidentiality | .50 |
| Met clients/patients outside your office of in their homes | .44 |
| Provided education for staff so that they could deal more effectively with clients/patients | .44 |
| Individualizing treatment recommendations, α = 5.84 | |
| Designed care recommendations for clients/patients that focus on their strengths | .70 |
| Recommended use of informal support sources to clients/patients (e.g. churches, schools, community elders, family networks) | .71 |
| Recommended use of formal support programs to clients/patients (e.g. AA, illness specific support groups) | .67 |
| Modified standard advice for care to what you knew that your clients/patients were likely able to achieve | .54 |
| Was a role model of healthily behavior for your clients/patients | .55 |
| Got more education or training yourself to enhance or broaden your skills | .57 |
| Suggested alternative or complementary approaches to care | .51 |
| De-emphasized status differences between yourself and your clients/patients | .44 |
| Safeguarding of confidentiality, α = 5.79 | |
| Did not report certain information about clients/patients to others | .73 |
| Did not enter certain sensitive information into records | .71 |
| Maintained sensitive information only in alternative records or files | .67 |
| Limited access by colleagues or staff to sensitive files or records | .65 |
| Used techniques such as using coded information or identifiers | .59 |
| Provided alternative entrances/exits into your practice site | .46 |
| Avoided interactions with clients/patients in public settings if you crossed paths with them | .41 |
| Discussed with clients/patients how to interact in public settings if you encounter each other | .40 |
| Broadening of provider expertise, α = 5.72 | |
| Referred your client/patient to another provider to obtain additional expertise | .79 |
| Discussed with clients/patients the limits of provider expertise, including your own, that exists within your community | .60 |
| Referred your clients/patients to another provider to assure confidentiality | .56 |
| Built a network of providers for referrals | .53 |
| Spent “extra” time talking with clients/patients about their problems or care | .53 |
| Had your staff spend extra time with clients/patients | .47 |
| Adjusting treatment style, α = 5.71 | |
| Adjusted your use of language to accommodate your clients/patients (including use of translators) | .60 |
| Talked with family or friends of clients/patients who can provide support or information | .59 |
| Talked with your clients/patients about their family or friends who can provide support or information | .55 |
| Adjusted your pace to a level that was comfortable for your clients/patients | .51 |
| Was available to clients/patients before or after hours | .30 |
Data analyses
To reduce the number of dependent variables, a principal components analysis with equamax rotation was conducted on the 37 items. Results revealed 8 factors with an eigenvalue equal to or greater than 1, accounting for 58.4% of the variance. Using scree plot and content analyses, the ideal solution was found to be 5 factors, accounting for 48.8% of the variance. These 5 factors can be labeled Integrating Community Resources, Individualizing Treatment Recommendations, Safeguarding Confidentiality, Broadening Provider Expertise, and Adjusting Treatment Style. Items that loaded on multiple factors were retained for the factor on which the item had the highest factor loading. Table 2 provides coeffcient alphas and item factor loadings.
Data analyses were based on participants’ community size, provider type, and state. For community size, we initially categorized participants into rural or urban for the random selection process. For this categorization, we used the definition provided by the Federal Office of Management and Budget.20 According to this definition, a metropolitan statistical area (MSA), as opposed to a non-metropolitan statistical area (non-MSA), is a geographic area consisting of a city with at least 50,000 residents or an urbanized area with a total population of at least 100,000 (75,000 in New England). However, because of criticisms in the literature regarding such rural-urban dichotomies,4,10,21 for community size, we used a different approach. We used cluster analyses based on contiguous population size to define community size for the participants’ communities. e resulting definition includes four distinct categories: urban, small urban, rural, and small rural. An urban community was defined as a geographic area containing a population of 35,000 people or more; a small urban community was defined as a geographic area containing a population of 15,000 to 34,999 people; a rural community was defined as a geographic area containing a population of 3,500 to 14,999 people; and a small rural community was defined a geographic area containing a population of fewer than 3,500 people.
For provider type, the eight health care provider groups were combined into two conceptually-related groups. Physicians, physician assistants, nurse practitioners, and registered nurses were combined into the group labeled physical health care providers; psychologists, psychiatrists, social workers, and mental health counselors were combined into a group labeled behavioral health care providers.
Based on these categorizations, the primary statistical analysis used was multivariate analyses of variance (MANOVA) with three independent variables of State (two levels: New Mexico and Alaska), Provider Type (two levels: physical health providers and behavioral health providers), and Community Size (four levels: urban, small urban, rural, small rural), and the five factors as the dependent variables. If the overall MANOVA was statistically significant, univariate analyses of variance and Duncan’s Range Test were conducted. Cohen’s d was used to determine the effect size for significant findings.
Results
Results of the 2 (State) X 2 (Provider Type) X 4 (Community Size) MANOVA revealed significant overall main effects for Community Size, F(15,4191) =10.99, p<.001, and Provider Type, F(5,1518) = 113.74, p<.001. As shown in Table 3, the main effect for State, and all two-way or three-way interactions did not reach statistical significance.
Table 3.
F-STATISTICS AND PROBABILITIES FOR THE MANOVAa
| Overall MANOVA | Integrating community resources | Individualizing treatment recommendations | Safeguarding confidentiality | Broadening provider expertise | Adjusting treatment style | |
|---|---|---|---|---|---|---|
| Community size | F(15,4191)=10.99, p<.001 | F(3,1522)=7.34, p<.001 | F(3,1522)=5.69, p<.001 | F(3,1522)=10.14, p<.001 | F(3,1522)=10.57, p<.001 | F(3,1522)=6.78, p<.001 |
| Provider type | F(5,1518)=113.74, p<.001 | F(1,1522)=96.65, p<.001 | F(1,1522)=179.28, p<.001 | F(1,1522)=281.61, p<.001 | F(1,1522)=11.18, p<.01 | F(1,1522)=27.22, p<.01 |
| State | F(5,1518)=2.52, p>.05 | F(1,1522)=4.03, p>.05 | F(1,1522)=.25, p>.05 | F(1,1522)=.47, p>.05 | F(1,1522)=.26, p>.05 | F(1,1522)=.85, p>.05 |
| Community size X Discipline | F(15,4191)=1.02, p>.05 | F(3,1522)=.83, p>.05 | F(3,1522)=1.43, p>.05 | F(3,1522)=.26, p>.05 | F(3,1522)=1.42, p>.05 | F(3,1522)=1.52, p>.05 |
| Community size X State | F(15,4191)=1.04, p>.05 | F(3,1522)=1.36, p>.05 | F(3,1522)=.18, p>.05 | F(3,1522)=.06, p>.05 | F(3,1522)=.84, p>.05 | F(3,1522)=.94, p>.05 |
| Provider type X State | F(15,4191)=.33, p>.05 | F(1,1522)=.01, p>.05 | F(1,1522)=.30, p>.05 | F(1,1522)=.12, p>.05 | F(1,1522)=.18, p>.05 | F(1,1522)=.92, p>.05 |
| Community size X Provider type X State | F(15,4191)=.70, p>.05 | F(3,1522)=1.13, p>.05 | F(3,1522)=.52, p=.38 | F(3,1522)=.35, p>.05 | F(3,1522)=.15, p>.05 | F(3,1522)=1.47, p>.05 |
Bold values represent significant findings.
Community size
Univariate analysis of variance revealed significant differences for Community Size on all five factors, namely, Integrating Community Resources, Individualizing Treatment Recommendations, Safeguarding Confidentiality, Broadening Provider Expertise, and Adjusting Treatment Style. All univariate results are provided in Table 3. For Integrating Community Resources, all four levels of Community Size differed significantly from one another. Table 4 provides means and standard deviations for all effects. As shown, providers in small rural communities were the most likely to integrate community resources to address perceived barriers to care, followed in decreasing order by providers in rural, small urban, and urban communities. e largest mean difference, between providers in small rural communities and providers in urban communities, reflected a medium effect size (Cohen’s d = .55). For Individualizing Treatment Recommendations, providers in urban communities were less likely to individualize treatment recommendations than providers in small urban, rural, and small rural communities, who did not differ from one another. e largest mean difference, between providers in small rural communities and providers in urban communities, reflected a very small effect size (Cohen’s d = .13). For Safeguarding Confidentiality, providers in urban communities were less likely to take special precautions to safeguard confidentiality than providers in small urban, rural, and small rural communities, who did not differ from one another. e largest mean difference between urban and small urban reflected a medium effect size (Cohen’s d = .56). e only other difference was between providers in rural and providers in small urban communities; however, this difference reflected a negligible effect size (Cohen’s d = .04).
Table 4.
MEANS AND STANDARD DEVIATIONS FOR FIVE ADAPTATIONS FACTORS BY COMMUNITY SIZE, PROVIDER TYPE, AND STATE
| Community
|
Provider type
|
State
|
|||||||
|---|---|---|---|---|---|---|---|---|---|
| Small rural
(n=147) |
Rural
(n=366) |
Small urban
(n=386) |
Urban
(n=643) |
Physical health
(n=850) |
Behavioral health
(n=692) |
Alaska
(n=722) |
New Mexico
(n=820) |
||
| Integrating community resources | M | 4.59 | 4.08 | 3.64 | 3.06 | 3.2 | 4.09 | 3.76 | 3.46 |
| SD | 1.91 | 1.81 | 1.97 | 1.88 | 1.92 | 1.89 | 1.97 | 1.94 | |
| Individualizing treatment recommendations | M | 6.70 | 6.70 | 6.67 | 6.48 | 6.04 | 7.28 | 6.63 | 6.57 |
| SD | 1.47 | 1.49 | 1.68 | 1.76 | 1.68 | 1.32 | 1.63 | 1.67 | |
| Safeguarding confidentiality | M | 2.75 | 2.71 | 2.80 | 2.42 | 1.85 | 3.56 | 2.66 | 2.58 |
| SD | 1.92 | 1.90 | 1.90 | 1.91 | 1.57 | 1.86 | 1.94 | 1.89 | |
| Broadening provider expertise | M | 5.57 | 5.23 | 4.96 | 4.73 | 5.13 | 4.81 | 5.02 | 4.95 |
| SD | 1.42 | 1.55 | 1.60 | 1.66 | 1.64 | 1.58 | 1.63 | 1.62 | |
| Adjusting treatment style | M | 6.61 | 6.21 | 6.11 | 6.03 | 5.94 | 6.4 | 6.1 | 6.19 |
| SD | 1.68 | 1.66 | 1.77 | 1.75 | 1.77 | 1.66 | 1.78 | 1.69 | |
For Broadening Provider Expertise, all four levels of Community Size differed significantly from one another. Providers in small rural communities were most likely to take measures to broaden their expertise in attempt to address barriers to care followed in descending order by providers in rural, small urban, and urban communities. e largest mean difference between small rural and urban providers reflected a medium effect size (Cohen’s d = .52). For Adjusting Treatment Style, providers in small rural communities were more likely to adjust their treatment styles than providers in rural, small urban, and urban communities, who did not differ significantly from one another. e largest mean difference between small rural and urban providers reflected a moderate effect size of (Cohen’s d = .38).
Provider type
Significant differences were revealed between the two provider types on all five factors. Behavioral health care providers were more likely than physical health care providers to integrate community resources (Cohen’s d = .67; medium effect size), individualize their treatment recommendations (Cohen’s d = .81; large effect size), take special measures to safeguard their clients’ confidentiality (Cohen’s d = 1.0; large effect size), and adjust their treatment (Cohen’s d = .27; small effect size). Physical health care providers were more likely than behavioral health care providers to take measures to broaden their expertise (Cohen’s d = .20; small effect size). e means for the Provider Type main effects are shown in Table 4.
Rank order of adaptations by community size
Based on average ratings, Table 5 provides the highest ranked of the 37 individual adaptation items, separated by rural and urban providers. Although being a role model for healthy behavior was the most common adaptation used by both rural and urban providers, considerable differences existed in the use of other adaptations. Rural providers were more likely to use adaptations that were associated with client interactions such as adjusting language, spending more time with clients, and minimizing a status difference between the client and provider. Urban providers were more likely to use adaptations that were associated with training, integrating familial support, and designing care recommendations around clients’ strengths.
Table 5.
RANK ORDER OF ADAPTATIONS MOST OFTEN REPORTED BY RURAL AND URBAN PROVIDERS
| Item | Rural ranking | Urban ranking |
|---|---|---|
| Was a role model of healthy behavior for your clients/patients | 1 | 1 |
| Adjusted your use of language to accommodate your clients/patients (including use of translators) | 2 | 6 |
| Spent “extra” time talking with clients/patients about their problems or care | 3 | 7 |
| De-emphasized status differences between yourself and your clients/patients | 4 | 5 |
| Got more education or training yourself to enhance or broaden your skills | 5 | 2 |
| Talked with your clients/patients about their family or friends who can provide support or information | 6 | 3 |
| Designed care recommendations for clients/patients that focus on their strengths | 7 | 4 |
| Recommended use of informal support sources to clients/patients (e.g. churches, schools, community elders, family networks) | 8 | 8 |
| Adjusted your pace to a level that was comfortable for your clients/patients | 9 | 10 |
| Modified standard advice for care to what you knew that your clients/patients were likely able to achieve | 10 | 11 |
| Recommended use of formal support programs to clients/patients (e.g. AA, illness specific support groups) | 11 | 9 |
Discussion
This large-scale survey of health care providers in rural and urban areas of Alaska and New Mexico revealed that rural and urban providers have significantly different approaches to dealing with barriers to ethical and effective health care delivery. When looking at the five clusters of adaptations to health care barriers, providers in rural communities compared with urban providers showed the greatest differences with regard to broadening provider expertise, followed by how they integrate community resources and safeguard confidentiality. Behavioral and physical health care providers differed significantly in the degree to which they endorsed adaptations. Following is a detailed discussion of each of the five adaptation clusters.
Integrating community resources
Providers’ community size was strongly related to the integration of community resources into service provision, with small rural providers using this set of adaptations the most, followed by rural, small urban, and urban providers. Specifically, the smaller the community, the more the providers consulted community leaders or elders regarding cultural issues, participated in community events, and integrated additional community resources. e smaller the communities, the more its providers have contact with the influential leaders of the community, which provides them the opportunity to discuss how best to support the community. Additionally, providers in smaller communities have much less anonymity than do urban providers; thus, if they do not participate in community events, their absence is noticed. It may well be that if providers in small communities do not attend community events, they would run the risk of being identified as outsiders, a label that in turn may have a negative impact on community members’ willingness to seek their services and to form a therapeutic alliance with him or her.
Provider type had a significant effect on the degree to which providers integrated community resources into their practice, with behavioral health care providers using this set of adaptations more often than physical health care providers. is may occur due to the nature of concerns discussed between behavioral health care providers and their clients. at is, often, behavioral health concerns are related to community level stressors, such as the loss of industry, environmental changes, or community level tragedies (such as, floods, fires, or suicides). Such presenting problems may lead behavioral health care providers to utilize available resources to supplement their treatment protocols. They may also lead to community efforts towards prevention.
Broadening provider expertise
Community size was significantly related to providers’ use of adaptations that serve to broaden their expertise, with small rural providers using this set of adaptations the most, followed by rural, small urban, and urban providers. is finding is consistent with previous research documenting the lack of specialization and opportunity for provider consultation in smaller communities.22,23 It is likely that the smaller the community, the greater the need for providers to become broader and more proficient in their health care skills. These providers also will have a greater need for a comprehensive network for referrals as they will be confronted with many issues outside their own scope of practice that will require assistance from specialists.
In addition to differences by community size, significant dissimilarity was revealed between provider types, with physical health care providers indicating a higher frequency of adaptations designed to expand their expertise than is found among behavioral health care providers. One possibility is that physical health care providers have more opportunity for providing referrals and engaging in more training. at is, access to referrals and care decrease as specialization increases, with a particular dearth of mental health or behavioral health care providers in rural areas.24 For example, a nurse practitioner may need to make a referral to a general practicing physician, who in turn may need to make a referral to a specialist (e.g., a cardiologist). In comparison, behavioral health care providers may be limited in both their available referral network and their need for specialists.
Safeguarding confidentiality
Community size had a significant relationship to the means through which health care providers safeguard patients’ confidential information. Urban providers rated such adaptations significantly lower than did providers from all other community sizes, which did not differ from each another. is finding lends support to the notion that concerns regarding confidentiality are more prevalent in smaller communities and pressingly require creative solutions. Confidentiality challenges in smaller communities arise in many contexts and providers recognize that the ensuing lack of anonymity and possibility for stigmatization greatly affects their patients. To ensure that patients are comfortable seeking services and feel that their privacy remains protected, providers in small communities take increased steps to safeguard confidentiality.
Provider type had a similar relationship with the safeguarding of client’s confidentiality, with behavioral health providers reporting a greater frequency than physical health providers of safeguarding confidentiality. Due to the stigmatization of mental illness, it is not surprising that behavioral health providers make extra efforts to maintain their clients’ confidentiality.
Individualizing treatment recommendations
Community size was not related to whether providers choose alternative treatment approaches as adaptations to barriers to health care. However, provider type did have a significant relationship with such utilization. Specifically, behavioral health care providers reported using alternative treatment approaches as adaptations to treatment barriers significantly more often than physical health care providers did. Often, behavioral health care requires a time-intensive effort to establish therapeutic alliance, a process that may be greatly enhanced by alternative treatment approaches. Alternatives may focus on clients’ strengths and abilities, as well as integrating informal and formal supports. Physical health care providers may engage in similar treatment approaches, but do so less frequently than their behavioral health care colleagues.
Adjusting treatment style
The finding that providers in small rural communities differed from those in all other geographic settings with regard to how they adjust their interactive style with their patients or clients is not surprising. Care providers in small communities must take great care to use natural helpers in their community and must adapt to the cultural styles of their patients. is may require a slowing down of language or an adjustment of office hours. It certainly appears to mean that friends and family members are more likely to become involved in a patient’s care than they would be in another setting. Behavioral and physical care providers were similar in this respect.
Implications
The fact that providers in rural communities adapt differently from how urban providers adapt in the face of barriers to care has implications for policy-makers, educators, and care providers in rural communities. First, funding agencies can provide adequate funds for the adaptations frequently used by rural health care providers. For example, adequate funding is necessary for rural providers as they seek out additional supervision or training given that they must treat or at least refer patient with presenting concerns outside of providers’ specialty. Second, the fact that rural providers must make major adaptations to meet the needs of their patients provides additional evidence of the need for more providers and, especially, more specialists in rural communities. Without addressing this need for more providers, patients in rural communities will continue to suffer the serious effects of ongoing health disparities. Technological advances and the use of telemedicine for distance service delivery will provide some relief from this problem, but are not cure-alls.
Third, these findings contribute to the body of knowledge that rural providers face circumstances that are quantitatively and qualitatively different than those faced by urban providers. Providers who will work in rural communities must secure specialized training in rural health care issues. These findings provide the basis for such training that should be integrated into established university and medical school training programs. These adaptations also point to the need for continuing education opportunities for rural providers, with adequate provision of backup services while providers leave their communities for such training. rough better preparation and ongoing training, the burn-out and rapid turnover of rural providers may be reduced, resulting in more stable and consistent rural health care services.
Limitations
Several limitations must be borne in mind when interpreting these findings. First, results are based on self-report and, as such, may not precisely reflect respondents’ actual behaviors. Second, the overall response rate was 52.2%, which means that nearly half of the potential participants declined to participate. However, given the fact that the individuals targeted were busy professionals and that the survey took up to an hour to complete, the response rate seems understandable; it is also relevant that response rates to surveys in general have been declining for at least two decades.25 Third, although a significant amount of time and effort was put into developing and refining the survey, including the use of expert consultants and extensive piloting efforts, no validity or reliability data have been collected thus far. Fourth, our community size classification schema is based on our data rather than other classifications schemas such as those established by the United States Department of Agriculture (http://www.ers.usda.gov/Briefing/Rurality/). However, given the particular population characteristics of Alaska and New Mexico, the established schemas would have not given us the opportunity to look at finer nuances of community size. Nevertheless, our classification schema corresponds closely to the USDA schemas and is more intuitively sound.
Conclusion
Health care providers in rural areas face a variety of barriers, including heterogeneity of rural inhabitants, economic impediments, geographic impediments (e.g., distance, travel, and weather), financial constraints, cultural differences, lack of access to specialty care, and confidentiality concerns. is study differentiated adaptations utilized by a diverse range of communities and provider types to overcome health care barriers. e study demonstrated that providers from small rural and rural communities are more likely than small urban and urban providers to find ways to integrate community resources, safeguard client confidentiality, and seek out additional expertise. In addition, behavioral health providers appear to integrate community resources, find alternative treatment approaches, and make efforts to safeguard confidentiality more than physical health providers, whereas physical health providers make more attempts or have more opportunities to broaden their expertise.
Acknowledgments
This research was supported by grant 1RO1DA13139 from the National Institute on Drug Abuse. Dr. Roberts also acknowledges the support of a Career Development Award (1KO2MH01918) from the National Institute of Mental Health. We gratefully acknowledge the contributions of Ginger Mongeau, Marcine Mullen, and David Neal at Behavioral Health Research and Services in Alaska; Pamela Monaghan Geernart, Alexis Kaminsky, Audrey Solimon, and Katherine Green Hammond in New Mexico; and Lee Sechrest and Patrick McKnight in Arizona.
Appendix—Recommendations for Improving Rural Health Care
System integration and flexibility
Integration of physical health care, mental health care, and substance abuse treatment under one roof
Development of a system that provides a single health care team, a single patient medical chart, and a single billing system
Collaboration across disciplines
Integration of social workers and mental health care providers into primary care practices for screening, consultation, and support
Development of flexibility through creative strategies such as mobile clinics, non-traditional clinic hours, and house calls
Community outreach
Creation of sustainable community alliances and collaborations
Establishment of networks of natural helpers, including clergy, families, neighbors, shelters, jails, peace officers, peer supports, schools, and so forth to make the most effective use of formal and informal local resources
Involvement, collaboration, and solicitation of local supports and through tapping into existing local strengths and values
Professional outreach
Good communication and regular contact with distant specialists who serve in consultant roles, perhaps using telemedicine technology
Self-awareness to know when to ask for help and when to initiate referrals
Measures to increase skills vis-à-vis identification and diagnosis, emergency care, competent referral
Development of rural outreach systems for mutual support and guidance and to reduce professional isolation
Cultural sensitivity and awareness
Community-specific health-care-planning with focus on partnership building
Responsiveness to the unique needs of each individual rural communities
Capacity for building trust with local residents
Willingness to engage in advocacy role that facilitates help-seeking
Reduction or elimination of miscommunication between the provider and consumer due to language or cultural barriers
Notes
- 1.Brems C, Johnson ME, Warner TD, et al. Barriers to healthcare as reported by rural and urban interprofessional providers. J Interprof Care. 2006 Mar;20(2):105–18. doi: 10.1080/13561820600622208. [DOI] [PubMed] [Google Scholar]
- 2.Ricketts TC. The rural patient. In: Geyman JP, Norris TE, Hart LG, editors. Textbook of rural medicine. Seattle, WA: McGraw-Hill; 2000. pp. 15–26. [Google Scholar]
- 3.DeLeon PH, Wakefield M, Hagglund KJ. The behavioral health care needs of rural communities in the 21st century. In: Stamm BH, editor. Rural behavioral health care: an interdisciplinary guide. Washington, DC: American Psychological Association; 2003. Feb, pp. 23–31. [Google Scholar]
- 4.Goins RT, Mitchell J. Health-related quality of life: does rurality matter? J Rural Health. 1999 Spring;15(2):147–56. doi: 10.1111/j.1748-0361.1999.tb00734.x. [DOI] [PubMed] [Google Scholar]
- 5.Roberts LW, Johnson ME, Brems C, et al. Ethical disparities: challenges encountered by multidisciplinary providers in fulfilling ethical standards in the care of rural and minority people. J Rural Health. 2007 Fall;23 Suppl:89–97. doi: 10.1111/j.1748-0361.2007.00130.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Geller JM, Muus KJ. The role of rural primary care physicians in the provision of mental health services. Journal of the Washington Academy of Science. 2000;86(3):131–42. [Google Scholar]
- 7.Mathews-Cowey S. Too much responsibility, too little help: family carers and the rural mental health care system. Journal of Family Studies. 2000;6:267–71. [Google Scholar]
- 8.Sammons MT. Nonphysician prescribers in rural settings: unique roles and opportunities for enhanced mental health care. In: Stamm BH, editor. Rural behavioral health care: an interdisciplinary guide. Washington, DC: American Psychological Association; 2003. pp. 121–31. [Google Scholar]
- 9.Stockdill JW, Ciarlo JA. Aging, mental illness, and the frontier. Journal of the Washington Academy of Science. 2000 December;86(3):107–15. [Google Scholar]
- 10.Rost K, Fortney J, Fischer E, et al. Use, quality, and outcomes of care for mental health: the rural perspective. Med Care Res Rev. 2002 Sep;59(3):231–65. doi: 10.1177/1077558702059003001. discussion 266–71. [DOI] [PubMed] [Google Scholar]
- 11.Sherrill WW, Crew L, Mayo RM, et al. Educational and health services innovation to improve care for rural Hispanic communities in the US. Educ Health (Abingdon) 2005 Nov;18(3):356–67. doi: 10.1080/13576280500312850. [DOI] [PubMed] [Google Scholar]
- 12.Smith LD, Peck PL, McGovern RJ. Knowledge and access barriers to counseling services in a rural population. Journal of Mental Health and Aging. 2002 Fall;8(3):183–95. [Google Scholar]
- 13.Wahl C, Gregoire JP, Teo K, et al. Concordance, compliance and adherence in health-care: closing gaps and improving outcomes. Healthc Q. 2005;8(1):65–70. doi: 10.12927/hcq..16941. [DOI] [PubMed] [Google Scholar]
- 14.Brems C, Johnson ME. Challenges and uniqueness of rural and frontier services in the United States. Journal of Psychological Practice. 2007;14:93–122. [Google Scholar]
- 15.Roberts LW, Battaglia J, Warner TD, et al. Stigma, ethics, and the frontier: challenges in caring for people with serious illnesses in Alaska and New Mexico. Arctic Research of the United States. 2001;15:2–13. [Google Scholar]
- 16.Johnson ME, Brems C, Roberts LW, et al. Ethics continuing education needs as reported by rural and urban mental healthcare providers. Prof Psychol Res Pr. 2006;37:183–9. [Google Scholar]
- 17.Roberts LW, Battaglia J, Epstein RS. Frontier ethics: mental health care needs and ethical dilemmas in rural communities. Psychiatr Serv. 1999 Apr;50(4):497–503. doi: 10.1176/ps.50.4.497. [DOI] [PubMed] [Google Scholar]
- 18.Dillman DA. Mail and internet surveys: the tailored design method. New York: John Wiley and Sons, Inc; 2000. pp. 149–93. Chapter 4. [Google Scholar]
- 19.Brems C, Johnson ME, Warner TD, et al. Exploring differences in caseloads of rural and urban healthcare providers in Alaska and New Mexico. Public Health. 2007 Jan;121(1):3–17. doi: 10.1016/j.puhe.2006.07.031. [DOI] [PubMed] [Google Scholar]
- 20.Office of Management and Budget. Federal Register. Part IV: notices. 1999 Oct 20;64(202):56628–44. Available at http://www.whitehouse.gov/omb/inforeg/msa-recommend.pdf.
- 21.Ciarlo JA, Zelarney PT. Focusing on “Frontier”: isolated rural America. Journal of the Washington Academy of Science. 2000 December;86(3):1–24. [Google Scholar]
- 22.Barbopoulos A, Clark JM. Practising psychology in rural settings: issues and guidelines. Canadian Psychol. 2003 Nov;44(4):410–24. [Google Scholar]
- 23.Rosenblatt RA. The health of rural people and the communities and environments in which they live. In: Geyman JP, Norris TE, Hart LG, editors. Textbook of rural medicine. Seattle, WA: McGraw-Hill; 2000. pp. 3–14. [Google Scholar]
- 24.Johnson ME, Brems C, Warner TD, et al. Rural-urban health care provider disparities in Alaska and New Mexico. Adm Policy Ment Health. 2006 Jul;33(4):504–7. doi: 10.1007/s10488-005-0001-7. [DOI] [PubMed] [Google Scholar]
- 25.Tourangeau R. Survey research and societal change. Annu Rev Psychol. 2004;55:775–801. doi: 10.1146/annurev.psych.55.090902.142040. [DOI] [PubMed] [Google Scholar]