PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Prog Community Health Partnersh. Author manuscript; available in PMC 2014 January 3.
Published in final edited form as:
PMCID: PMC3879950
NIHMSID: NIHMS526859

Patients’ and Community Leaders’ Perceptions Regarding Conducting Health Behavior Research in a Diverse, Urban Clinic Specializing in Rheumatic Diseases

Abstract

Background

Disparities in the incidence, prevalence, severity, care, and outcomes for rheumatic diseases exist among racial and ethnic groups compared with White Americans.

Objective

This paper describes a community-based participatory research (CBPR) approach engaging researchers, community leaders, and patients in purposeful dialogues related to the implementation of health behavior research in an urban rheumatic disease clinic.

Methods

Seven focused discussions were led in either English or Spanish. Discussions were audiotaped and transcribed verbatim.

Results

Six community leaders and nine patients participated in the seven scheduled focused discussions. Transcripts uncovered five major themes that assisted with study design: trust, patient–provider relationship, study implementation suggestions, decreased functional capacity, and access to healthcare.

Conclusions

Engaging community partners and patients in informal and formal discussions from early phases of research design through implementation, followed by systematic application of these insights, may serve to accelerate the potential for translation from findings into improved clinical practice and optimal outcomes.

Keywords: Community health partnerships, community-based participatory research, community health research, health disparities, rheumatic diseases, musculoskeletal diseases, urban population

Rheumatic diseases are among the most common health problems in the United States.1,2 Marked differences in the incidence, prevalence, severity, processes of care, and outcomes in a number of rheumatic conditions exist among racial and ethnic groups as compared with White Americans. Furthermore, access to health care and treatment also differs by community. Rheumatic diseases are a diverse group of disorders, with several shared characteristics. Among the rheumatic diseases, rheumatoid arthritis and osteoarthritis are the most frequent, but serious and potentially life-threatening disorders such as systemic lupus erythematosus and scleroderma are also included. These diseases have a long course and can affect many aspects of a patient’s life. Among the effects are impairments in activities of daily living, occupational and social functioning, and cognitive ability.3

Data from the Behavioral Risk Factor Surveillance System indicates that people with arthritis have significantly worse health-related quality of life and report more than twice as many unhealthy days as well as three times as many days with activity limitations than those without arthritis.4 Additionally, arthritis is strongly associated with major depression, possibly through its role in creating functional limitations.5

Examining the racial/ethnic disparities in the prevalence and impact of arthritis is important to identifying priority populations for community-based public health interventions.6 To examine these differences in the prevalence and impact of arthritis, the U.S. Centers for Disease Control and Prevention (CDC) combined data from the 2002, 2003, and 2006 National Health Interview Surveys and then analyzed racial/ethnic differences. After adjusting for age, gender, and body mass index, results indicate that arthritis disproportionately affects racial/ethnic minorities. Arthritis-attributable activity limitation, arthritis-attributable work limitation, and severe joint pain were higher for non-Hispanic Blacks, Hispanics, and multiracial or other respondents with arthritis compared with non-Hispanic Whites with arthritis.7

In a CDC analysis of the 2002 National Health Interview Survey, where racial and ethnic differences in the prevalence and impact of arthritis were examined, when asked about joint pain (excluding the neck or back), nearly one in four (24.6%) adult respondents with arthritis reported severe joint pain during the preceding 30 days.8 Blacks with physician-diagnosed arthritis had a higher prevalence of severe pain attributable to arthritis, compared with Whites (34% vs. 23%). A greater proportion of Hispanics with physician-diagnosed arthritis reported severe joint pain, compared with Whites (33% versus 23%).8 Although important clinical and psychosocial differences have been reported between Hispanic and non-Hispanic patients with regard to rheumatoid arthritis and systemic lupus erythematosus, there are few published data on the prevalence or characteristics of other rheumatic diseases in the U.S. Hispanic population.9

Despite efforts to quantify the prevalence and impact of rheumatic diseases, it is difficult to disentangle the reasons for racial and ethnic disparities for specific diseases, including arthritis. The underlying causes span the range of possibilities from genetic predisposition to environmental exposures to behavioral and social etiologies.10 Attention is focused on the identification of potentially modifiable environmental and social/behavioral factors behind racial and ethnic disparities in disease manifestations and outcome.10

This paper describes a formative, qualitative study using focused discussions and a CBPR approach to engage researchers, community leaders from a health partnership program (HPP), and predominantly Hispanic and African-American patients in purposeful dialogues focused on the implementation and dissemination of health behavior research in an urban rheumatic disease clinic. Six community leaders and nine patients participated in seven different discussions. These discussions were the second phase of a three-phased study entitled Health Beliefs and Health Practices Among Minorities with Rheumatic Disease (NCT00069342). The first phase of the study involved the validation of patient reported outcomes, including an inventory to assess participants’ complementary and alternative medicine (CAM) use through cognitive interviews.11 Once the cognitive interview phase of the study was complete, community leaders from an HPP and patients attending an urban community clinic were recruited to participate in focused discussions to evaluate and provide recommendations for the implementation of phase three of the study which involved the quantitative evaluation of patient reported outcomes over a 6-month period.

The community leaders included in the first and last of the focused discussions were part of the HPP, a collaborative effort between the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) and Washington, DC, area community organizations and representatives.12 The HPP was initially proposed to engage community leaders in addressing health disparities in arthritis and other rheumatic diseases and promoted community-based medical research by focusing on public health education, patient care, access to clinical investigations, and recruitment to research careers. This partnership was developed with the goal of opening a NIAMS ambulatory urban clinic site from which to launch research exploring the natural history of rheumatic diseases. Previous to the HPP, researchers had visibility through professional ties to advocacy organizations such as the National Arthritis Foundation and the National Fibromyalgia Association; however, overall community visibility and knowledge of the National Institutes of Health remained low in this urban community. HPP partners included local government agencies, schools, universities, faith-based organizations, civic and community groups, voluntary and professional organizations, and private businesses. These partners offered input on the HPP’s direction and helped promote NIAMS research, training, and education resources. The NIAMS Community Health Center (CHC), located in the Unity Health Care, Inc.’s Cardozo Clinic in northwest Washington, DC, was part of the HPP and served as the platform for its research, education, and training activities. All patients at the CHC were enrolled in the Natural History of Rheumatic Diseases in Minorities study, an observational study designed to gather information on how minority populations are disproportionately affected by rheumatic diseases. This study was approved by the NIAMS Institutional Review Board. Participants provided written informed consent to participate in this study.

Before scientific and human subjects review of this study, community leaders and patients who attended the clinic were asked to provide feedback in the design of the health behavior and health beliefs study. It was during collaborative meetings that the community leaders emphasized the importance of conducting a focused discussion phase to the study to gain insight into the potential study participants and the community at large. This type of CBPR is an appropriate approach to provide researchers and clinicians with insights into the local context of a community, including information about policies or norms of conduct, as well as methodological considerations, including best practices for recruitment and retention of participants and venues in which to conduct the research.13,14 Community partnerships often allow researchers to gain access to communities where there are real or perceived barriers to recruitment of participants through a collaborative approach to generating new research hypotheses and culturally relevant interventions.15 These participatory strategies that develop relationships with the community being studied are often characterized as “reflexive, flexible, and iterative,” thus offering substantial opportunities to engage the communities as active contributors while simultaneously exploring local knowledge and perceptions.16 When patients are also included in this participatory process, researchers gain an “insider’s” perspective into the nature of their disease as well as insight into the relevancy of variables selected for assessment and proposed outcome measures.17 Additionally, CBPR approaches such as those we will describe help rebuild and maintain trust in vulnerable communities with past history of being analyzed, stereotyped and exploited by outside groups. Since trust is vital in the success of any research projects in these communities, it is critical that all groups or partners take action to build and maintain trust.18

To better understand the community as a whole, the community leaders and the clinic patients served by a federally funded, outpatient specialty clinic for rheumatic diseases, our objectives for the formative study phase described in this paper included identifying the following from both community leaders’ and patients’ perspectives.

  1. Potential facilitators and barriers in conducting community health research in minority patients with rheumatic diseases and to develop strategies that could facilitate a study that explored health beliefs and behaviors among minority patients with rheumatic diseases.
  2. Best methods for assessing health beliefs and behaviors including CAM practices in minority patients with rheumatic diseases.

METHODS

Sample

Seven focused discussion were conducted in either English (five) or Spanish (two) with each group led by one bilingual moderator and at least one facilitator (Table 1). Six community leaders and nine patients for a total sample size of 15 participated in the seven scheduled focused discussions. Two of the seven discussions included community leaders and were held as the first and last of the seven discussions so that the investigators could explore any additional themes raised during the five patient discussions.

Table 1
Participant Demographics

Data Collection

A moderator led each discussion according to a Moderator’s Guide that included preselected questions and probes (Table 2). The participants were encouraged to ask questions or further develop discussions spontaneously. The questions were grouped in categories recommended by Krueger19: Opening, introductory, transition, key, and closing. The facilitator took notes, made observations, and managed the equipment (tape recorder, microphones, etc.). Moderators with similar racial and ethnic, cultural, and linguistic backgrounds as the participants were selected and trained to conduct the focused discussions. The selection of concordant moderators was essential to facilitate communication and identify cultural nuances.11

Table 2
Focused Discussion Questions

Data Analysis

Focused discussions were taped and transcribed verbatim. In addition to the transcripts of the recordings, the facilitator’s field notes were used to verify and provide additional contextual information supporting the transcribed data. Transcript-based analysis of the data was considered to be the most rigorous choice for analyzing the information generated during the discussions and was undertaken by two independent reviewers.20 Field notes taken by the facilitator during the discussions and debriefing sessions with the moderators and facilitators were also examined. The general analysis plan included the following: identifying the major themes, considering the choice and meaning of the words, considering the context, and observing for the consistency of responses. A thematic analysis was conducted by an independent evaluator to decrease the potential for investigator bias followed by a second independent evaluator who conducted a qualitative computer analysis utilizing NVivo Version 6.0 as an additional tool to index and cross-reference the thematic analyses. The independent thematic analyses were then verified and consensus was met with the agreement of the original discussion moderator and facilitators.

RESULTS

Thematic analysis of the focused discussions yielded five major themes by community members and patients that assisted in guiding subsequent study design and implementation: trust, patient–provider relationship, study implementation suggestions, issues surrounding decreased functional capacity and access to care (Tables 3 and and4).4). The following section will be divided into community leaders’ perspectives followed by patients’ perspectives with illustrative quotes presented in Table 3 and Table 4, respectively.

Table 3
Community Leaders’ Focused Discussion Themes and Illustrative Quotes
Table 4
Patients’ Focused Discussion Themes and Illustrative Quotes

Community Leaders’ Perspectives (Table 3)

Trust

Community leaders provided the research team with their perspectives regarding potential trust issues related to conducting research in their community. The ethical issues surrounding the U.S. Public Health Service Tuskegee Study, such as mistrust in researchers and being used as “guinea pigs,” were spontaneously raised during the discussions. Community leaders also offered suggestions for facilitating research in their community, such as the need for researchers to be visible and involved when it came to community activities beyond research. The importance of including church leaders in the research process and the importance of using racially concordant interviewers to optimize trust was also discussed.

Trust was discussed in the context of the researchers’ community involvement. Community leaders felt it was important for the research team to be visible to the community beyond the focus of the research implementation. Leaders explained that individuals need to perceive a direct benefit for themselves, their family or the community at large.

Church leaders were described by the community leaders as individuals who could be trusted in the community and should be sought after for endorsement and support of research activities within their church community. It should be noted that the HPP included leaders from faith-based organizations in the Black and Hispanic communities. Community leaders also raised the issue of researcher and subject concordance. They emphasized that trust could be enhanced through racial/ethnic and language concordance.

Patient–Provider Relationship

It has been suggested that doctors can dominate patient–provider interactions simply because they may be more knowledgeable than patients.21 However, shared decision-making in healthcare is ethical22 and has been shown to lead to better health outcomes.23,24

During the discussions, community leaders conveyed their views surrounding the importance of a good patient–provider relationship when eliciting answers about patients’ health behaviors, including CAM use. In addition to spending time with patients, community leaders emphasized the need for patients’ impressions of their provider to be positive and friendly.

Study Implementation Suggestions

In congruence with the themes of patient–provider relationships, communication, and trust, community leaders unanimously indicated that face-to-face interviews would be the best method for data collection when compared with paper-and-pencil, web-based, or mailed surveys. Community leaders also suggested allowing participants to be accompanied by a person of their choosing to the face-to-face interviews to reduce their apprehension with the interview process. Invaluable suggestions regarding the length and complexity of the questionnaires, particularly the inventory for CAM practices, were incorporated in an effort to reduce respondent burden. The consensus among the community leaders was that the shorter and more concise the interview questionnaires, the better.

Decreased Functional Capacity

Community leaders discussed functional capacity among the rheumatic disease patients in terms of potential changes in self-care. They emphasized the need for researchers and providers to understand the contextual environment in which the patients live and their responsibilities of daily living. They also mentioned lack of knowledge of technical health definitions among patients and the need for communicating clear healthcare and behavior change options. These recommendations could include options such as “exercise,” which admittedly may sound daunting to patients with low mobility.

Access to Health Care

In referring to their healthcare experience, community leaders expressed the importance of understanding that in a growing environment of managed care, the one-size-fits-all, in-and-out physician visit does not meet patients’ needs. The lack of access to rheumatology specialty clinics was also addressed as a barrier. Furthermore, the NIAMS CHC was identified as one of the only sources for rheumatology specific care for those living in this community.

Patients’ Perspectives (Table 4)

Trust

In general, trust was less of a concern for patients than it was for community leaders. Rather than bringing up historical events that could potentially reduce the trust they feel with their providers, patients who attended the NIAMS CHC spoke of trust as a core requirement upon which their relationship with their provider was built. One issue raised was the fear of switching doctors owing to discomfort with any type of change.

Patient–Provider Relationship

Overall, patients conveyed satisfaction with their current situation with regard to patient–provider relations. Patients described the importance of feeling that they were more than just “a patient” with the opportunity to take time to discuss their illness. For at least one patient, the fact that her physician knew and remembered her mother and sister was extremely important. The existence of two-way interactions with their provider made patients feel valued.

Study Implementation Suggestions

In terms of implementation strategies, patient participants reminded the research team of the importance of being cognizant of respondents’ level of discomfort during the interview process based on their own pain and symptom experiences, which would make sitting for longer than a half an hour nearly impossible without additional pain and stiffness. Some patients felt that 90-minute interviews would be too much in one sitting and recommended taking short breaks. Patients also requested that interviews be held in a location that is comfortable and in general did not feel like a doctor’s office.

Decreased Functional Capacity

Patients conveyed frustration with regard to limited ability to perform everyday tasks, sometimes in a cyclic pattern. One emerging theme was physical limitations causing interrupted or reduced work schedules and the financial ramifications for the patient and family. Some patients also spoke of how decreased functional capacity and decreased independence contributed to emotional suffering and depression.

Access to Health Care

Patients expressed feelings of being completely on their own in navigating what they believe is a confusing healthcare system. Some patients also identified the clinic as a sole source of guidance for their care.

LESSONS LEARNED

One of the most important lessons learned during this process of community engagement through collaborations with community leaders and patients was that of community-research “fit.” After reading the text and quotations in this paper one cannot help but notice that both community leaders and patients expressed their views of research within the context of their health status and their access to healthcare. The fact that this partnership was initiated by a team conducting clinical research at the National Institutes of Health potentially added further to issues surrounding program sustainability over time. How would the data be used? Would the community benefit directly from the research? Approaching the relationship as fragile from its inception provided the community with the time needed to evaluate the fit for the proposed research.

For this community, like many others who have experienced disparate health outcomes, and particularly when it comes to the burdens of chronic disease, engaging in research must have more immediate applicability and translation into improved individual and community outcomes for it to be embraced as a true partnership. By meeting informally at “town hall”–type gatherings for updates followed by a more formal focused discussion process, we were able to anticipate potential barriers and facilitators for the design and implementation of the third phase of our study. When specific elements of the study did not seem to “fit,” our research team needed to be open to changing our approach. These changes toward a better fit ranged from how we structured the planned discussions (community members suggested that we include them in the first and last groups rather than including only patients) to what types of incentives would be appreciated by the research participants (they suggested jar openers and pens designed especially for individuals with functional disabilities).

Including patients in the research partnership from the design phase of the study was instrumental to our success. Whereas the community leaders helped to guide our research implementation process with suggestions from how to approach the clinic patients to best methods for collecting data, the patients also offered us a “window” into their lives as individuals with a chronic illness and what was important to them such as having access to specialty care and understanding how to navigate complex health systems.

The community leaders, particularly African-American leaders, voiced more concerns about issues of trust and research participation than the patients did. Although the community leaders expressed concerns related to the infamous Tuskegee Study, where African-American men with syphilis did not understand the purpose of the study nor did they realize that they were deprived of effective treatment for the disease when penicillin became available in the 1950s,25 they also offered suggestions for increasing trust both in the community and with individual research participants. Their readiness to provide suggestions and guidance for improving trust in research seemed to imply the fact that in this context research was recognized as having potential to improve health in their community. These underlying beliefs informed the community leaders’ willingness to contribute to the success and value of the proposed research study.

In addition to their strong belief that researchers need to be consistently engaged and visible in the community, our community and patient partners raised the importance of researcher and subject concordance in the context of building trust. This view is supported by previous studies of patient–provider concordance where, if the provider was of the same ethnicity/race as the patient, the patient would be more satisfied and more likely to use healthcare services.26 Blacks were more satisfied with their provider, more likely to receive preventive care, and more likely to receive necessary medical care when the provider was Black. Although Hispanic patients paired with Hispanic physicians did not differ in their physician satisfaction ratings, they were more satisfied with their overall health care, but only if there was someone Hispanic in the provider’s office.26 This concept of the importance of concordance has also been described in participants’ decisions to enroll in research studies.27

CONCLUSION

Engaging community partners in informal and formal discussions from the early phases of research design through implementation, followed by systematic application of these insights, may serve to accelerate the potential for translation from findings into improved clinical practice and ultimately optimal patient and community outcomes. Clinical research partnerships between communities, providers, and patients are fragile and require vigilance and ongoing communication, as well as transparency. From the earliest stages, the research team must be willing to engage the community in a collaborative research design and work toward a community research “fit.” Community leaders and patients were able to provide valuable insight into barriers and facilitators such as improved accessibility to participation that could be adjusted to improve the “fit” and thus improve the quality and value of the information collected. Without this fit, each of the steps in the research process runs the risk of stalling and gaining momentum in the already difficult research to translation continuum is further compromised. Our research team continues to engage patient and community leaders as partners in research as we move from exploratory health behavior studies to self-care intervention studies.

Acknowledgments

The authors gratefully acknowledge the interdisciplinary collaboration and support of Gregory Dennis, MD, Margarita Velarde, CRNP, Kelli Carrington, MA CHES, Reva C. Lawrence, MPH, Janet Austin, PhD, Blakely Denkinger, RD, Madeline Michael, MS, RD, Robert Miranda-Acevedo, BA, Nicole Schuett, BS, Juan Mendoza, BS, patient-participants, and members of the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) Health Partnership Program (HPP). This work was supported by the intramural research program at the National Institutes of Health, Clinical Center and by NIAMS.

References

1. Helmick CG, Felson DT, Lawrence RC, et al. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part I. Arthritis Rheum. 2008;58(1):15–25. [PubMed]
2. Lawrence RC, Felson DT, Helmick CG, et al. Estimates of the prevalence of arthritis and other rheumatic conditions in the United States. Part II. Arthritis Rheum. 2008 Jan;58(1):26–35. [PMC free article] [PubMed]
3. Brown SC, Glass JM, Park DC. The relationship of pain and depression to cognitive function in rheumatoid arthritis patients. Pain. 2002 Apr;96(3):279–284. [PubMed]
4. Mili F, Helmick CG, Moriarty DG. Health-related quality of life among adults reporting arthritis: analysis of data from the Behavioral Risk Factor Surveillance System, US 19961999. J Rheumatol. 2003 Jan;30(1):160–166. [PubMed]
5. Dunlop DD, Lyons JS, Manheim LM, et al. Arthritis and heart disease as risk factors for major depression: the role of functional limitation. Med Care. 2004 Jun;42(6) [PubMed]
6. Beaton DE, Wright JG, Katz JN, et al. Development of the QuickDASH: Comparison of three item-reduction approaches. J Bone Jt Surg (Am) 2005 May;87(5): 1038–1046. [PubMed]
7. Bolen JL, Schieb L, Hootman JM, et al. Differences in the prevalence and severity of arthritis among racial/ethnic groups in the United States, National Health Interview Survey, 2002, 2003, and 2006. Prev Chronic Dis. 2010 May;7(3):A64. [PMC free article] [PubMed]
8. CDC-MMWR. Racial/ethnic differences in the prevalence and impact of doctor-diagnosed arthritis—United States, 2002. MMWR Morb Mortal Wkly Rep. 2005;54(5):119–123. [PubMed]
9. Escalante A, del Rincon IMD. Epidemiology and impact of rheumatic disorders in the United States Hispanic population. Curr Opin Rheumatol. 2001 Mar;13(2):104–110. [PubMed]
10. Hoovestol RA, Mikulis TR. Environmental exposures and rheumatoid arthritis risk. Curr Rheumatol Rep. 2011 Oct;13(5):431–439. [PubMed]
11. Wallen GR, Middleton KR, Rivera-Goba MV, et al. Validation of English and Spanish-language patient reported outcome measures in underserved rheumatic disease patients. Arthritis Res Ther. 2011 Jan 5;13(1):R1. [PMC free article] [PubMed]
12. Grady C, Hampson LA, Wallen GR, et al. Exploring the ethics of clinical research in an urban community. Am J Public Health. 2006 Nov;96(11):1996–2001. [PubMed]
13. Bogart LM, Uyeda K. Community-based participatory research: Partnering with communities for effective and sustainable behavioral health interventions. Health Psych. 2009 Jun;28(4):391–393. [PMC free article] [PubMed]
14. Peterson JW, Dobie S. Community’s voice in managing childhood asthma: The assessment phase of research. J Health Care Poor Underserved. 2005 Nov;16(4):747–759. [PubMed]
15. Hyatt RR, Gute DM, Pirie A, et al. Transferring knowledge about human subjects protections and the role of Institutional Review Boards in a community-based participatory research project. Am J Public Health. 2009 Nov;99:S526–531. [PubMed]
16. Cornwall A, Jewkes R. What is participatory research? Soc Sci Med. 1995 Dec;41(12):1667–1676. [PubMed]
17. Brown M. Perspectives on outcome: What disability insiders and outsiders each bring to the assessment table. Am J Phys Med Rehabil. 2009 Nov;90(11):S36–40. [PubMed]
18. Christopher S, Watts V, McCormick AK, et al. Building and maintaining trust in a community-based participatory research partnership. Am J Public Health. 2008 Aug;98(8):1398–1406. [PubMed]
19. Krueger RA. Developing questions for focus groups. Thousand Oaks, CA: Sage; 1998.
20. Krueger RA. Analyzing & reporting focus group results. Thousand Oaks, CA: Sage; 1998.
21. Turner B. Medical power and social knowledge. London: Sage publications; 1987.
22. Edwards M, Davies M, Edwards A. What are the external influences on information exchange and shared decision-making in healthcare consultations: A meta-synthesis of the literature. Patient Educ Counsel. 2008 Apr;75(1):37–52. [PubMed]
23. Adams RJ, Smith FJ, Ruffin RE. Impact of the physician’s participatory style in asthma outcomes and patient satisfaction. Ann Allergy Asthma Immunol. 2001 Mar;86:263–271. [PubMed]
24. Kaplan SH, Greenfield S, Gandek B, et al. Characteristics of physicians with participatory decision-making styles. Ann Intern Med. 1996 Mar 1;124(5):497–504. [PubMed]
25. Liu M, Davis K. Lessons from a horse named Jim: A clinical trials manual from the Duke Clinical Research Institute. New York: Taylor & Francis Group; 2001.
26. Saha S, Komaromy M, Koepsell TD, et al. Patient-physician racial concordance and the perceived quality and use of health care. Arch Intern Med. 1999 May 10;159(9):997–1004. [PubMed]
27. Rivera-Goba MV, Dominguez DC, Stoll P, et al. Exploring decision-making of HIV-infected Hispanics and African Americans participating in clinical trials. J Assoc Nurses AIDS Care. 2011 Jul-Aug;22(4):295–306. [PMC free article] [PubMed]