- Email: [email protected]
Patient-provider relationship as mediator between adult attachment and self-management in primary care patients with multiple chronic conditions
Accepted Manuscript Patient-provider relationship as mediator between adult attachment and self-management in primary care patients with multiple chronic conditions
Katja Brenk-Franz, Bernhard Strauss, Fabian Tiesler, Christian Fleischhauer, Nico Schneider, Jochen Gensichen PII: DOI: Reference:
S0022-3999(17)30019-3 doi: 10.1016/j.jpsychores.2017.04.007 PSR 9320
To appear in:
Journal of Psychosomatic Research
Received date: Revised date: Accepted date:
9 January 2017 23 March 2017 14 April 2017
Please cite this article as: Katja Brenk-Franz, Bernhard Strauss, Fabian Tiesler, Christian Fleischhauer, Nico Schneider, Jochen Gensichen , Patient-provider relationship as mediator between adult attachment and self-management in primary care patients with multiple chronic conditions. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Psr(2017), doi: 10.1016/ j.jpsychores.2017.04.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Title Patient-Provider Relationship as Mediator between Adult Attachment and SelfManagement in Primary Care Patients with Multiple Chronic Conditions
RI
PT
Katja Brenk-Franz (PhD) 1,2; Bernhard Strauss (PhD) 2; Fabian Tiesler (MSc) 1,2; Christian Fleischhauer (MD) 1; Nico Schneider (MSc) 1; Jochen Gensichen (MD, MSc, MPH) 1,3
SC
1 Institute of General Practice and Family Medicine, Jena University Hospital 2 Institute of Psychosocial Medicine and Psychotherapy, Jena University Hospital
NU
3 Institute of General Practice, Ludwig-Maximilians-University Munich
MA
CORRESPONDING AUTHOR: Katja Brenk-Franz
D
Institute of Psychosocial Medicine and Psychotherapy, University Hospital Jena,
PT E
Stoystrasse 3, 07740 Jena, Germany, [email protected]
CE
phone +49-3641 9 35 49 0
AC
fax +49-3641 9 36 54 6
Funding acknowledgement The APRICARE study is funded by the German Research Society, DFG: GE 2073 / 5-1. 1
ACCEPTED MANUSCRIPT Abstract
PT
Objective: The conceptual model of attachment theory has been applied to understand the predispositions of patients in medical care and the patient-provider relationship. In patients with chronic conditions insecure attachment was connected to poorer self-management. The patient-provider relationship is associated with a range of health related outcomes and self-management skills. We determined whether the quality of the patient-provider relationship mediates the link between adult attachment and self-management among primary care patients with multiple chronic diseases.
NU
SC
RI
Method: 209 patients with a minimum of three chronic diseases (including type II diabetes, hypertension and at least one other chronic condition) between the ages of 50 and 85 from eight general practices were included in the APRICARE cohort study. Adult attachment was measured via self-report (ECR-RD), self-management skills by the FERUS and the patientprovider relationship by the PRA-D. The health status and chronicity were assessed by the GP. Multiple mediation analyses were used to examine whether aspects of the patientprovider relationship (communication, information, affectivity) are a mediators of associations between adult attachment and self-management.
D
MA
Results: The analysis revealed that the quality of the patient-provider relationship mediated the effect of attachment on self-management in patients with multiple chronic conditions. Particularly the quality of communication and information over the course of treatment has a significant mediating influence.
CE
PT E
Conclusion: A personalized, attachment-related approach that promotes active patientprovider communication and gives information about the treatment to the patient may improve self-management skills in patients.
AC
Keywords: Attachment, Physician-patient-relationship, Primary Care, Self-management, Mediation analysis
2
ACCEPTED MANUSCRIPT INTRODUCTION Self-management has a key part as an essential element of evidence-based medical care for patients with chronic diseases in primary care [1]. But there are different definitions. Selfmanagement is a complex, multi-component construct that includes aspects such as coping and selfefficacy. We defined it as the ability of patients to manage their problems actively and independently and to pursue their goals. Primary health care is the first point of contact for health care for
AC
CE
PT E
D
MA
NU
SC
RI
PT
most people. It is mainly provided by GPs (general practitioners). A major task in primary care is the treatment of elderly patients with multiple chronic diseases [2, 3]. The GP is often the main contact for these patients. However, to implement effective patient-centered selfmanagement programs in primary care, it is important to understand the differences in patient interaction styles. Attachment theory provides a psychosocial model to explain the individual differences in experience and behavior in relation to interpersonal closeness and distance to stress and affect regulation in situations that are subjectively perceived as threatening [4, 5]. Based on the central assumptions of attachment theory, a model for the activation of the attachment system in adulthood was developed [6]. The model supports the adoption of the attachment activation by chronic diseases [7]. There are other model assumptions which describe the different influences of insecure attachment on the maintenance of diseases or chronic disease through example low self-management skills [8]. Researchers found evidence in diabetic patients, that insecure attachment was associated with poorer diabetes self-management and negative outcomes [9, 10]. In our own studies, we also found a relationship between insecure attachment and low self-management in patients with diabetes and other chronic conditions. Attachment anxiety was significantly linked to impaired coping and lower self-efficacy, hope, dietary control, and physical activity. Attachment avoidance, on the other hand, was associated with lower levels of social support and health-care use [11]. The present study is based on a conceptual classification of the two attachment dimensions anxiety and avoidance [12]. Patients scoring high on the avoidance scale have learned to suppress their attachment needs and use deactivating attachment strategies. The main characteristics of deactivating strategies are denying attachment needs, trivializing risks, as well as the repression of negative emotions and cognitions [13]. Avoidant patients are also characterized by high levels of self-reliance, greater interpersonal distance [14] and minimal help seeking behavior and expression of distress [15, 16]. They cope with stress by cognitive distancing from their emotions, denial and distraction [17]. Among the hyper-activating strategies, patients with higher scores of anxiety attachment show increased attention to threatening situations with their chronically activated attachment system leading to the perception of more risks (e.g. increased awareness of symptoms of illness reported to the physician) and to a hyperactive search for proximity, e.g. by the catastrophizing of symptoms [13]. Patients´ attachment styles can influence their health seeking behavior and the ability to accept help from primary care physicians and other health care professionals [18]. Some studies show the importance of the family physician as a primary caregiver [19]. Physicians are powerful figures who provide emphatic support by 3
ACCEPTED MANUSCRIPT listening and care in phases of activation of the attachment system [20]. Accordingly, we hypothesized that characteristic of the patient-provider relationship (especially the GPpatient relationship) has a mediating influence of the relationship between attachment characteristics of patients with multiple chronic diseases in primary care and their selfmanagement. METHOD
PT
Study Design and Recruitment
PT E
D
MA
NU
SC
RI
The APRICARE study (Adult Attachment in Primary Care study, funded by the German Research Society, DFG: GE 2073 / 5-1) included 219 patients from eight general practices in Germany. It was designed as a multi-center, prospective, longitudinal, observational cohort study. In each practice, a list of patients was drawn up based on the electronic database of the GP. This list comprised all patients who had three predefined chronic diseases (type II diabetes, hypertension, and at least one other chronic condition out of a standardized list of chronic diseases [21]), who were between the ages of 50 and 85 and had consulted their GP at least once within the previous 3-month period. Emergency patients, patients from other family practices and those unable to give informed consent were excluded from the study. A total of 25 to 35 patients from each of the eight units were included in the study. The recruitment was carried out in accordance with the primary care research recruitment rules of "The German Multi-Care-study" [21]. Recruitment and baseline data collection took place from March 2012 until June 2012. The study was conducted in accordance with the "Declaration of Helsinki", the guidelines of Good Clinical Practice and was approved by the Institutional Review Board of the Jena University Hospital (No.3009-12/10). Data Collection
AC
CE
25 to 35 of the eligible patients from the list of multi-morbidity patients with the predefined chronic diseases were selected at random (by a list of random numbers) and invited to participate in the study by means of a telephone call or a letter from the practices. Patients willing to participate and who agreed to the terms of the study were asked to sign an informed consent form at the practice. The physician then filled out a sheet with the basic documentation and inclusion criteria; the patient received a comprehensive questionnaire for self-assessment, which had to be completed at home and returned in a sealed envelope to the practice. Patients were reimbursed for this, receiving EUR 10 for their efforts. The questionnaires the physicians had to complete included the ICD-10 diagnosis and assessment of the severity of the participants’ chronic conditions. Each GP received EUR 500 per measurement point. Measures
4
ACCEPTED MANUSCRIPT The patients’ socioeconomic status was assessed based on the recommendations of the Association of "Epidemiologic Methods" in the German Association of Epidemiology [22]. State of health and documentation of multi-morbidity
RI
PT
In order to determine the patients’ health status and multi-morbidity various scores were used. The patients completed the list of chronic diseases and assessed their state of health by means of a visual analogue scale from 0 to 100 [23] German: [24]. The physician also documented the chronic diseases based on the equivalent list of chronic diseases. Moreover, the degree of chronicity was rated using the Cumulative Illness Rating Scale for Geriatrics. The CIRS-G is a multi-morbidity index based on disease severity grouped at organ system levels [25]. A 4-level classification of severity is used to assess the 14 organ systems [26].
SC
Attachment
PT E
D
MA
NU
A self-rating questionnaire was used to measure adult attachment. The Experience of Close Relationships – Revised in its German version [12, 27] is a dimensional measure of adult attachment style on the two subscales - avoidance and anxiety - with a seven-point Likert scale. In general, individuals with higher scores of attachment avoidance report lower intimacy. They find intimacy uncomfortable and prefer to seek their independence. Patients scoring highly on the attachment anxiety dimension have a tendency to fear rejection and abandonment. A requirement to the instrument was to put specific focus on the relationship to the current or previous partner. The validation of the German version showed Cronbach's alpha reliability scores of .91 and .92 for the two relevant sub-scales. Due to the participants being elderly and having several ailments, the instrument had to be brief and easy to use and understand, as well as easily implemented in primary care, so we used the short form ECR-RD12 [28].
CE
Self-Management
AC
A self-management questionnaire (FERUS26) for the measurement of resources and selfmanagement skills was used [29]. The FERUS26 includes 26 items to be answered on a fivepoint Likert scale. The subscales self-verbalization, coping, self-efficacy and hope can be combined to the scale "self-management skills". The internal consistency (Cronbach's alpha) for the subscales ranged between 0.86 -0.93. Patient-Provider Relationship The patients' perception of the patient-provider relationship was measured by the Patient Reaction Assessment (PRA) [30, 31]. The three subscales include communication, information and affectivity, which were measured on a seven-point Likert scale (higher scores indicate higher quality of the patient-provider relationship). The subscale information describes the perception of the patient to see the doctors as transmitters of information, who provide a more detailed description of the diseases and treatment. Communication 5
ACCEPTED MANUSCRIPT subscale assesses the ability to actively participate in the communication process. The affective subscale measures a respectful interaction of the physician with the patient and if the physician shows emotional support and understanding. The internal consistency is high (between .87 and .91). A total sum score can be calculated for determining the quality of physician-patient relationship. Cronbach´s alpha for the total score ranged between .83 -.91 in former studies [30, 31]. Statistical Analysis
AC
CE
PT E
D
MA
NU
SC
RI
PT
Assessment of the potential mediation effect of the patient-provider medical relationship relationship was evaluated using the Simple and Multiple Mediation Procedure for SPSS by Preacher and Hayes [32]. The hypothesized mediation model guiding this analysis is illustrated in Figure 1. To begin with, the criteria for a possible mediation can be tested. I) the association of the primary independent variable (attachment dimension: avoidance or anxiety) with the potential mediator (physician-patient relationship) must be significant (i.e. a ≠ 0 in the figure); furthermore II) the association of the primary independent variable (attachment dimension: avoidance or anxiety) with the dependent variable (selfmanagement) must be significant (i.e. c ≠ 0); and III) the mediator (physician-patient relationship) must be significantly associated with the dependent variable controlling for the primary independent variable (i.e. b ≠ 0). If these conditions are met and the association between the independent variable and the dependent variable is significantly reduced with the inclusion of the potential mediator in the model (i.e. c’ is statistically smaller than c) a significant mediation is assumed. The significance of the mediation effect was tested using the Simple Mediation Procedure by Preacher and Hayes [32]. The influence of the relevant sociodemographic variables and health status were controlled. An alpha level of p ≤ 0.05 was used for tests of statistical significance. Statistical analysis was performed using IBM SPSS 23.0 for Windows (Chicago, IL, USA). Furthermore, multiple mediation analyses also according to Preacher and Hayes [32] were calculated to examine whether the subscales of the PRA-D (communication, interaction and affectivity) also mediated the relationship between the attachment dimensions (anxiety and avoidance) and self-management skills. Although the number of missing values was very low, we used procedures for the management of missing values [33]. Multiple imputation procedures are commonly seen as the most adequate method for dealing with missing values in complex data sets [34]. SPSS uses a Markov-Chain-Monte-Carlo algorithm known as fully conditional specification (FCS) or chained equations imputation [35]. Multiple imputation was used to generate 10 sets of ‘‘complete’’ data with no missing values.
Figure 1 Conceptual model: the mediation effect of patient-provider relationship on attachment and self-management in older patients with multimorbidity 6
SC
RI
PT
ACCEPTED MANUSCRIPT
RESULTS
MA
NU
The socio-demographic data of the study participants are described in Table 1. A total of 219 patients (95 females) were included in the study. Patients' ages ranged from 50 to 85, with a mean age of 66.4 years ± 8.3 (Table 1). The self-reported health status ranged from 0 to 100, with a mean of 6.4 (SD = 2.1). Table 1. Characteristics of the apricare-sample (N=219)
PT E
AC
CE
Marital status
Education
Categories 50-61 62-73 74-85 Female Male Married Single Divorced Widowed Middle school Secondary modern school High school Missing Min Max
D
Variables Age M = 66.4 SD = 8.3 Sex
Number of Chronic Diseases (Physician) Multimorbidity Index (CIRS-G) Adult Attachment/ Avoidance
Frequency 67 110 42 95 124 158 9 18 34 63 112 37 7
Percentage 30.6 50.2 19.2 43.4 56.6 72.1 4.1 8.2 15.6 28.9 51.4 17.0 2.7 Mean
SD
3
18
6.4
2.5
2
38
8.3
4.2
1
7
1.9
1.3
7
ACCEPTED MANUSCRIPT
6.2
2.3
1.4
26
85
62.6
11.8
45
105
90.4
12.6
5 5 12
35 35 35
26.6 31.4 32.4
8.5 4.6 4.0
PT
1
RI
Adult Attachment/ Anxiety Self-Management and Resources (total) Patient-Provider Relationship (PRA-D total) PRA-D Subscales Communication Information Affectivity
PT E
D
MA
NU
SC
Using the Simple Mediation Procedure for SPSS by Preacher and Hayes (2004) we found that the relationship between the predictor (attachment related anxiety) and the criterion (selfmanagement skills) can be mediated through the quality of the patient-provider relationship) (Figure 2). All direct paths showed significant correlations. The path coefficient for the indirect effect remains statistically significant, therefore a partial mediation can be assumed. The confidence interval for the indirect effect ranged from -1.55 to -0.40. The overall model is considered significant Z = -3.33, p <0.001. For the model with the predictor attachment-related avoidance, the criterion self-management skills and the mediator patient-provider relationship, we found the relationships reflected in Figure 3. The confidence interval for the indirect effect ranges from -1.04 to -0.13, the path coefficient for the indirect effect is Bc' = -0.96; SEc' = 0.59 (ns), so that there is a complete mediation. The overall model is statistically significant (Z = -2.52; p <0.05).
AC
CE
Figure 2 The mediation effect of patient-provider relationship on anxiety and selfmanagement
Figure 3 The mediation effect of patient-provider relationship on anxiety and selfmanagement
8
ACCEPTED MANUSCRIPT
MA
NU
SC
RI
PT
Multiple Mediation analysis [32] considering the subscales of the PRA-D suggesting that the relationship between anxiety and self-management skills is mediated through the PRA-D subscales information and communication. The standardized regression coefficient for the indirect effect with simultaneous testing of all three subscales of the PRA-D is βc = -1.4 and is statistically significant (see Figure 4). Accordingly a partial mediation can be assumed. The overall model reached significance R2 = 0.16, p <0.001. Control variables were age, gender and the degree of chronicity (CIRS-G). In the model of avoidance and self-management only communication could be identified as a mediator. The standardized regression coefficient for the indirect effect is not statistically significant (βc´ = -0.95). We assumed full mediation (see Figure 5). The overall model is significant, R2 = 0.15, p <0.001.
AC
CE
PT E
D
Figure 4 The multivariate mediation effect of patient-provider relationship on anxiety and self-management
Figure 5 The multivariate mediation effect of patient-provider relationship on avoidance and self-management 9
RI
PT
ACCEPTED MANUSCRIPT
SC
DISCUSSION
AC
CE
PT E
D
MA
NU
The results of the mediation analyses supported the hypothesis that the patient-provider relationship mediates the relation between adult attachment and self-management in patients with multi-morbidity. We were also able to show that the relation between attachment-related anxiety and self-management was mediated by the subscales information and communication of the PRA-D and the relation between attachment-related avoidance and self-management was mediated by the subscale communication as parts of the patient-provider medical relationship. The communication scale reflects the patients’ experience in initiating communication with the GP concerning some aspect of their illness and treatment. It was confirmed that communication is one of the most important aspects in the patient-provider relationship [36, 37]. Therefore, it is not surprising that communication (such as the ability to ask questions during treatment) also has an influence on the relationship between patient characteristics and self-management in anxiety and in avoidant patients. This is consistent with the results of Ciechanowksi et al. (2001) who also identified the subscale communication as an influential factor in diabetes patients with avoidant attachment on health-related outcomes [38]. However, the information transfer to patients is an important task for physicians, and is the basis for behavioral change in selfmanagement [39]. In our study, the subscale information measured if patients had been given information about the possible side effects of their treatment/medication, and whether the treatment plan had been presented to them transparently for the following weeks and months. This could have an influence on the self-management skills of patients with anxious attachment, but only long-term studies or trials can answer this more precisely. Patients with anxious attachment do not feel resilient and resistant by the chronic activation of the attachment system [40]. They are often prone to vigorously searching the proximity of the other [41] , catastrophizing in representation of symptoms [42] and exaggerating their expression of needs to get support from their caregiver [43].Particularly patients with anxious attachment need clear, concise information about the course of their treatment 10
ACCEPTED MANUSCRIPT
RI
PT
this can enhance their self-management skills. In contrast, avoidant patients are characterized by high levels of self-reliance, greater interpersonal distance [14], as well as minimal help seeking behavior and expression of distress [15, 16]. They cope with stress with cognitive distancing from their emotions, denial and distraction [17], and seek information about their disease more independently of the GP as they tend to distrust others, expect them to be hostile, exploitative and not sensitive [44, 45]. In contrast, the subscale affectivity could not be identified as a mediator between the attachment-related anxiety and avoidance and self-management. This might be due to methodological limits of this subscale (e.g. a low internal consistency, alpha = 0.70, cf. [31]). On the other hand, the construct of emotional support cannot clearly be separated from the constructs of communication and information in the context of the doctor-patient-provider relationship because parts of it are already included.
MA
NU
SC
Patient surveys from a European Community study "European Project on Patient Evaluation of General Practice Care" (EUROPEP) have published the five most desirable qualities they wish for in a GP, i.e. willingness to answer questions clearly and concisely and to listen, quick accessibility, and confidentiality of patient data, information, and sufficient consultation time [46]. None of the requests explicitly or exclusively mention the affective components of the patient-provider relationship. Nevertheless, we think it is an important factor, but future research should examine the influence of the affective component (respectful interactions with the patient, emotional support and understanding) in more detail.
AC
CE
PT E
D
It has already been proven, that the individual attachment style of the patients can influence their self-care skills and behavior [9-11]. Additionally, we were able to find indications in this study that the quality of the patient-provider relationship can mediate this association. In order to improve the self-management in patients with anxiety attachment, the practitioner should give more structured information about methods of self-treatment and the side effects of medication. For these patients, the GP often is the main point of contact. In contrast, patients with avoidant attachment seek information about treatment, selftreatment and side effects more externally in books or on the internet [47]. If physicians reflect the effects of different attachment styles, they can provide positive and beneficial relationships and communication styles in different patients with chronic conditions [13, 44].
11
ACCEPTED MANUSCRIPT REFERENCES
AC
CE
PT E
D
MA
NU
SC
RI
PT
1. E.H. Wagner, B.T. Austin, C. Davis, M. Hindmarsh, J. Schaefer, A. Bonomi, Improving chronic illness care: translating evidence into action, Health Aff (Millwood) 20 (2001) 64-78. 2. WONCA Europe Council, The European definition of general practice / family medicine. http://www.woncaeurope.org/sites/default/files/documents/Definition%203rd%20ed%202 011%20with%20revised%20wonca%20tree.pdf, 2002 (accessed 12.12.16). 3. M. Huber, J.A. Knottnerus, L. Green, H. van der Horst, A.R. Jadad, D. Kromhout, B. Leonard, K. Lorig, M.I. Loureiro, J.W. van der Meer, P. Schnabel, R. Smith, C. van Weel, H. Smid, How should we define health?, BMJ 343 (2011) d4163, doi: 10.1136/bmj.d4163. 4. J. Bowlby, A Secure Base. Clinical Applications of Attachment Theory, Routledge, London, 1988. 5. W.H. Berman, L. Marcus, E.R. Berman, Attachment in marital relations, in: M.B. Sperling, W.H. Berman (Eds.), Attachment in adults: theory, assessment and treatment. Guilford Press, New York, 1994. 6. P.R. Shaver, M. Mikulincer, What do self-report attachment measures assess?, in: W.S. Rholes, J.A. Simpson (Eds.), Adult attachment: theory, research, and clinical implications. Guilford Press, New York, 2004, pp. 17-54. 7. B. Strauss, K. Brenk-Franz, The relevance of attachment theory in medical care, in: J.J. Hunter, R.G. Maunder (Eds.), Improving patient treatment with attachment theory: A guide for primary care practitioners and specialists. Springer, 2016. 8. R.G. Maunder, J.J. Hunter, Attachment and psychosomatic medicine: developmental contributions to stress and disease, Psychosom Med 63 (2001) 556-567. 9. P. Ciechanowski, J. Russo, W. Katon, M. Von Korff, E. Ludman, E. Lin, G. Simon, T. Bush, Influence of patient attachment style on self-care and outcomes in diabetes, Psychosom Med 66 (2004) 720-728, doi: 10.1097/01.psy.0000138125.59122.23. 10. P.S. Ciechanowski, J.E. Russo, W.J. Katon, E.A. Walker, Attachment theory in health care: the influence of relationship style on medical students' specialty choice, Med Educ 38 (2004) 262-270, doi: 1767 [pii]. 11. K. Brenk-Franz, B. Strauss, F. Tiesler, C. Fleischhauer, P. Ciechanowski, N. Schneider, J. Gensichen, The influence of adult attachment on patient self-management in primary care the need for a personalized approach and patient-centred care, PLoS One 10 (2015) e0136723, doi: 10.1371/journal.pone.0136723. 12. R.C. Fraley, N.G. Waller, K.A. Brennan, An item response theory analysis of self-report measures of adult attachment, J Pers Soc Psychol 78 (2000) 350-365. 13. D. Thompson, P.S. Ciechanowski, Attaching a new understanding to the patientphysician relationship in family practice, J Am Board Fam Pract 16 (2003) 219-226. 14. J.G. Allen, H. Stein, P. Fonagy, J. Fultz, M. Target, Rethinking adult attachment: a study of expert consensus, Bull Menninger Clin 69 (2005) 59-80, doi: 10.1521/bumc.69.1.59.62266. 15. W.S. Rholes, J.A. Simpson, J.G. Stevens, Attachment orientations, social support, and conflict resolution in close relationships, in: J.A. Simpson, W.S. Rholes (Eds.), Attachment theory and close relationships. Guilford Press, New York, 1998, pp. 166-188.
12
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
16. M. Main, Avoidance in the service of attachment: a working paper, in: K. Immelman, G. Barlow, M. Main, L. Petrinovitch (Eds.), Behavioral development: the Bielefeld Interdisciplinary Project. Cambridge University Press, New York, 1981, pp. 651–693. 17. J.A. Feeney, Adult attachment, coping style and health locus of control as predictors of health behavior, Aust J Psychol 47 (1995) 171-177. 18. J.J. Hunter, R.G. Maunder, Using attachment theory to understand illness behavior, Gen Hosp Psychiatry 23 (2001) 177-182, doi: S0163-8343(01)00141-4 [pii]. 19. H.B. Frederiksen, J. Kragstrup, B. Dehlholm-Lambertsen, Attachment in the doctorpatient relationship in general practice: a qualitative study, Scand J Prim Health Care 28 (2010) 185-190, doi: 10.3109/02813432.2010.505447. 20. G. D'Elia, [Attachment theory: a biological basis for the patient-physician relation?], Lakartidningen 97 (2000) 1956-1959. 21. L. Simmons, N.J. Baker, J. Schaefer, D. Miller, S. Anders, Activation of patients for successful self-management, J Ambul Care Manage 32 (2009) 16-23, doi: 10.1097/01.JAC.0000343120.07844.a9. 22. K.H. Jöckel, B. Babitsch, B.M. Bellach, K. Bloomfield, J.H.P. Hoffmeyer-Zlontnik, J. Winkler, C. Wolf, [Recommendations of the working group "Epidemiologic Methods" in the German Association Epidemiology of GMDS and DGSMP to measure and quantify the sociodemographic characteristics in epidemiological studies], in: W. Ahrens, B.M. Bellach, K.H. Jöckel (Eds.), [Measurement of socio-demographic characteristics in epidemiology]. MMVVerlag, Munich, 1998, pp. 7-38. 23. The EuroQol Group, Euro-Qol - a new facility for the measurement of health-related quality of live, Health Policy 16 (1990) 199-208. 24. J.-M. Graf von der Schulenburg, C. Claes, W. Greiner, A. Uber, [The German version of the EuroQol questionnaire], Z f Gesundheitswiss 6 (1998) 3-20. 25. M.D. Miller, C.F. Paradis, P.R. Houck, S. Mazumdar, J.A. Stack, A.H. Rifai, B. Mulsant, C.F. Reynolds, 3rd, Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale, Psychiatry Res 41 (1992) 237248, doi: 0165-1781(92)90005-N [pii]. 26. C. Hudon, M. Fortin, H. Soubhi, Abbreviated guidelines for scoring the Cumulative Illness Rating Scale (CIRS) in family practice, J Clin Epidemiol 60 (2007) 212, doi: 10.1016/j.jclinepi.2005.12.021. 27. J.C. Ehrenthal, U. Dinger, A. Lamla, B. Funken, H. Schauenburg, [Evaluation of the German version of the attachment questionnaire "Experiences in Close Relationships-Revised" (ECR-RD)], Psychother Psychosom Med Psychol 59 (2009) 215-223, doi: 10.1055/s2008-1067425. 28. K. Brenk-Franz, J. Ehrenthal, T. Freund, N. Schneider, B. Strauss, F. Tiesler, H. Schauenburg, J. Gensichen, Evaluation of the Short Form of “Experience in Close Relationships” (Revised, German Version “ECR-RD12”) - a Tool to Measure Adult Attachment in Primary Care PLoS One (submitted). 29. M. Jack, FERUS - Questionnaire for Measuring Resources and Self-management skills, Hogrefe, Göttingen, 2007. 30. J.P. Galassi, R. Schanberg, W.B. Ware, The Patient Reactions Assessment: a brief measure of the quality of the patient-provider medical relationship, Psychol Assess 4 (1992) 346-351. 13
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
31. K. Brenk-Franz, G. Hunold, J.P. Galassi, F. Tiesler, W. Herrmann, T. Freund, C. SteurerStey, S. Djalali, A. Sönnichsen, N. Schneider, J. Gensichen, [Quality of the physician-patient relationship – evaluation of the German version of the Patient Reactions Assessment (PRAD)], Z Allg Med 92 (2016) 103-108. 32. K.J. Preacher, A.F. Hayes, Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models, Behav Res Methods 40 (2008) 879891. 33. J.L. Schafer, Analysis of incomplete multivariate data, London, Chapman & Hall, 1997. 34. D.B. Rubin, Multiple imputation after 18+ years (with discussion), JASA (1996) 473489. 35. W.R. Gilks, S. Richardson, D.J. Spiegelhalter, Markov Chain Monte Carlo in Practice, Chapman and Hall, London, 1996. 36. L.M. Ong, J.C. de Haes, A.M. Hoos, F.B. Lammes, Doctor-patient communication: a review of the literature, Soc Sci Med 40 (1995) 903-918. 37. K. Laemmel, [Interview in medicine], Praxis (Bern 1994) 85 (1996) 863-869. 38. P.S. Ciechanowski, W.J. Katon, J.E. Russo, E.A. Walker, The patient-provider relationship: attachment theory and adherence to treatment in diabetes, Am J Psychiatry 158 (2001) 29-35. 39. M. Lipkin, Jr., Physician-patient interaction in reproductive counseling, Obstet Gynecol 88 (1996) 31S-40S. 40. P.J. Meredith, J. Strong, J.A. Feeney, Evidence of a relationship between adult attachment variables and appraisals of chronic pain, Pain Res Manag 10 (2005) 191-200. 41. M. Mikulincer, P.R. Shaver, Attachment in adulthood: structure, dynamics, and change, Guilford, New York, 2007. 42. P. Ciechanowski, M. Sullivan, M. Jensen, J. Romano, H. Summers, The relationship of attachment style to depression, catastrophizing and health care utilization in patients with chronic pain, Pain 104 (2003) 627-637, doi: S0304395903001209 [pii]. 43. A. Slade, Attachment theory and research: implications for the theory and practice of individual psychotherapy with adults, in: J. Cassidy, P.R. Shavers (Eds.), Handbook of attachment: theory, research, and clinical applications. Guilford, New York, 1999. 44. P. Ciechanowski, W.J. Katon, The interpersonal experience of health care through the eyes of patients with diabetes, Soc Sci Med 63 (2006) 3067-3079, doi: 10.1016/j.socscimed.2006.08.002. 45. M. Dozier, K.L. Cue, L. Barnett, Clinicians as caregivers: role of attachment organization in treatment, J Consult Clin Psychol 62 (1994) 793-800. 46. A. Klingenberg, O. Bahrs, J. Szecsenyi, [How do patients evaluate general practice? German results from the European Project on Patient Evaluation of General Practice Care (EUROPEP)], Z Arztl Fortbild Qualitatssich 93 (1999) 437-445. 47. J.J. Hunter, R.G. Maunder, Foundations. Introduction, in: J.J. Hunter, R.G. Maunder (Eds.), Improving patient treatment with attachment theoy: A guide for primary care practitioners and specialists. Springer, 2016.
14
ACCEPTED MANUSCRIPT
Highlights
Adult attachment is associated with self-management of patients with chronic conditions. The quality of the patient-provider relationship mediated the link between adult attachment and self-management among primary care patients with multiple chronic diseases.
PT
Particularly the quality of communication and information over the course of treatment has a significant mediating influence.
AC
CE
PT E
D
MA
NU
SC
RI
An active patient-provider communication and gives information about the treatment to the patient may improve self-management skills in patients.
15
Katja Brenk-Franz, Bernhard Strauss, Fabian Tiesler, Christian Fleischhauer, Nico Schneider, Jochen Gensichen PII: DOI: Reference:
S0022-3999(17)30019-3 doi: 10.1016/j.jpsychores.2017.04.007 PSR 9320
To appear in:
Journal of Psychosomatic Research
Received date: Revised date: Accepted date:
9 January 2017 23 March 2017 14 April 2017
Please cite this article as: Katja Brenk-Franz, Bernhard Strauss, Fabian Tiesler, Christian Fleischhauer, Nico Schneider, Jochen Gensichen , Patient-provider relationship as mediator between adult attachment and self-management in primary care patients with multiple chronic conditions. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Psr(2017), doi: 10.1016/ j.jpsychores.2017.04.007
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Title Patient-Provider Relationship as Mediator between Adult Attachment and SelfManagement in Primary Care Patients with Multiple Chronic Conditions
RI
PT
Katja Brenk-Franz (PhD) 1,2; Bernhard Strauss (PhD) 2; Fabian Tiesler (MSc) 1,2; Christian Fleischhauer (MD) 1; Nico Schneider (MSc) 1; Jochen Gensichen (MD, MSc, MPH) 1,3
SC
1 Institute of General Practice and Family Medicine, Jena University Hospital 2 Institute of Psychosocial Medicine and Psychotherapy, Jena University Hospital
NU
3 Institute of General Practice, Ludwig-Maximilians-University Munich
MA
CORRESPONDING AUTHOR: Katja Brenk-Franz
D
Institute of Psychosocial Medicine and Psychotherapy, University Hospital Jena,
PT E
Stoystrasse 3, 07740 Jena, Germany, [email protected]
CE
phone +49-3641 9 35 49 0
AC
fax +49-3641 9 36 54 6
Funding acknowledgement The APRICARE study is funded by the German Research Society, DFG: GE 2073 / 5-1. 1
ACCEPTED MANUSCRIPT Abstract
PT
Objective: The conceptual model of attachment theory has been applied to understand the predispositions of patients in medical care and the patient-provider relationship. In patients with chronic conditions insecure attachment was connected to poorer self-management. The patient-provider relationship is associated with a range of health related outcomes and self-management skills. We determined whether the quality of the patient-provider relationship mediates the link between adult attachment and self-management among primary care patients with multiple chronic diseases.
NU
SC
RI
Method: 209 patients with a minimum of three chronic diseases (including type II diabetes, hypertension and at least one other chronic condition) between the ages of 50 and 85 from eight general practices were included in the APRICARE cohort study. Adult attachment was measured via self-report (ECR-RD), self-management skills by the FERUS and the patientprovider relationship by the PRA-D. The health status and chronicity were assessed by the GP. Multiple mediation analyses were used to examine whether aspects of the patientprovider relationship (communication, information, affectivity) are a mediators of associations between adult attachment and self-management.
D
MA
Results: The analysis revealed that the quality of the patient-provider relationship mediated the effect of attachment on self-management in patients with multiple chronic conditions. Particularly the quality of communication and information over the course of treatment has a significant mediating influence.
CE
PT E
Conclusion: A personalized, attachment-related approach that promotes active patientprovider communication and gives information about the treatment to the patient may improve self-management skills in patients.
AC
Keywords: Attachment, Physician-patient-relationship, Primary Care, Self-management, Mediation analysis
2
ACCEPTED MANUSCRIPT INTRODUCTION Self-management has a key part as an essential element of evidence-based medical care for patients with chronic diseases in primary care [1]. But there are different definitions. Selfmanagement is a complex, multi-component construct that includes aspects such as coping and selfefficacy. We defined it as the ability of patients to manage their problems actively and independently and to pursue their goals. Primary health care is the first point of contact for health care for
AC
CE
PT E
D
MA
NU
SC
RI
PT
most people. It is mainly provided by GPs (general practitioners). A major task in primary care is the treatment of elderly patients with multiple chronic diseases [2, 3]. The GP is often the main contact for these patients. However, to implement effective patient-centered selfmanagement programs in primary care, it is important to understand the differences in patient interaction styles. Attachment theory provides a psychosocial model to explain the individual differences in experience and behavior in relation to interpersonal closeness and distance to stress and affect regulation in situations that are subjectively perceived as threatening [4, 5]. Based on the central assumptions of attachment theory, a model for the activation of the attachment system in adulthood was developed [6]. The model supports the adoption of the attachment activation by chronic diseases [7]. There are other model assumptions which describe the different influences of insecure attachment on the maintenance of diseases or chronic disease through example low self-management skills [8]. Researchers found evidence in diabetic patients, that insecure attachment was associated with poorer diabetes self-management and negative outcomes [9, 10]. In our own studies, we also found a relationship between insecure attachment and low self-management in patients with diabetes and other chronic conditions. Attachment anxiety was significantly linked to impaired coping and lower self-efficacy, hope, dietary control, and physical activity. Attachment avoidance, on the other hand, was associated with lower levels of social support and health-care use [11]. The present study is based on a conceptual classification of the two attachment dimensions anxiety and avoidance [12]. Patients scoring high on the avoidance scale have learned to suppress their attachment needs and use deactivating attachment strategies. The main characteristics of deactivating strategies are denying attachment needs, trivializing risks, as well as the repression of negative emotions and cognitions [13]. Avoidant patients are also characterized by high levels of self-reliance, greater interpersonal distance [14] and minimal help seeking behavior and expression of distress [15, 16]. They cope with stress by cognitive distancing from their emotions, denial and distraction [17]. Among the hyper-activating strategies, patients with higher scores of anxiety attachment show increased attention to threatening situations with their chronically activated attachment system leading to the perception of more risks (e.g. increased awareness of symptoms of illness reported to the physician) and to a hyperactive search for proximity, e.g. by the catastrophizing of symptoms [13]. Patients´ attachment styles can influence their health seeking behavior and the ability to accept help from primary care physicians and other health care professionals [18]. Some studies show the importance of the family physician as a primary caregiver [19]. Physicians are powerful figures who provide emphatic support by 3
ACCEPTED MANUSCRIPT listening and care in phases of activation of the attachment system [20]. Accordingly, we hypothesized that characteristic of the patient-provider relationship (especially the GPpatient relationship) has a mediating influence of the relationship between attachment characteristics of patients with multiple chronic diseases in primary care and their selfmanagement. METHOD
PT
Study Design and Recruitment
PT E
D
MA
NU
SC
RI
The APRICARE study (Adult Attachment in Primary Care study, funded by the German Research Society, DFG: GE 2073 / 5-1) included 219 patients from eight general practices in Germany. It was designed as a multi-center, prospective, longitudinal, observational cohort study. In each practice, a list of patients was drawn up based on the electronic database of the GP. This list comprised all patients who had three predefined chronic diseases (type II diabetes, hypertension, and at least one other chronic condition out of a standardized list of chronic diseases [21]), who were between the ages of 50 and 85 and had consulted their GP at least once within the previous 3-month period. Emergency patients, patients from other family practices and those unable to give informed consent were excluded from the study. A total of 25 to 35 patients from each of the eight units were included in the study. The recruitment was carried out in accordance with the primary care research recruitment rules of "The German Multi-Care-study" [21]. Recruitment and baseline data collection took place from March 2012 until June 2012. The study was conducted in accordance with the "Declaration of Helsinki", the guidelines of Good Clinical Practice and was approved by the Institutional Review Board of the Jena University Hospital (No.3009-12/10). Data Collection
AC
CE
25 to 35 of the eligible patients from the list of multi-morbidity patients with the predefined chronic diseases were selected at random (by a list of random numbers) and invited to participate in the study by means of a telephone call or a letter from the practices. Patients willing to participate and who agreed to the terms of the study were asked to sign an informed consent form at the practice. The physician then filled out a sheet with the basic documentation and inclusion criteria; the patient received a comprehensive questionnaire for self-assessment, which had to be completed at home and returned in a sealed envelope to the practice. Patients were reimbursed for this, receiving EUR 10 for their efforts. The questionnaires the physicians had to complete included the ICD-10 diagnosis and assessment of the severity of the participants’ chronic conditions. Each GP received EUR 500 per measurement point. Measures
4
ACCEPTED MANUSCRIPT The patients’ socioeconomic status was assessed based on the recommendations of the Association of "Epidemiologic Methods" in the German Association of Epidemiology [22]. State of health and documentation of multi-morbidity
RI
PT
In order to determine the patients’ health status and multi-morbidity various scores were used. The patients completed the list of chronic diseases and assessed their state of health by means of a visual analogue scale from 0 to 100 [23] German: [24]. The physician also documented the chronic diseases based on the equivalent list of chronic diseases. Moreover, the degree of chronicity was rated using the Cumulative Illness Rating Scale for Geriatrics. The CIRS-G is a multi-morbidity index based on disease severity grouped at organ system levels [25]. A 4-level classification of severity is used to assess the 14 organ systems [26].
SC
Attachment
PT E
D
MA
NU
A self-rating questionnaire was used to measure adult attachment. The Experience of Close Relationships – Revised in its German version [12, 27] is a dimensional measure of adult attachment style on the two subscales - avoidance and anxiety - with a seven-point Likert scale. In general, individuals with higher scores of attachment avoidance report lower intimacy. They find intimacy uncomfortable and prefer to seek their independence. Patients scoring highly on the attachment anxiety dimension have a tendency to fear rejection and abandonment. A requirement to the instrument was to put specific focus on the relationship to the current or previous partner. The validation of the German version showed Cronbach's alpha reliability scores of .91 and .92 for the two relevant sub-scales. Due to the participants being elderly and having several ailments, the instrument had to be brief and easy to use and understand, as well as easily implemented in primary care, so we used the short form ECR-RD12 [28].
CE
Self-Management
AC
A self-management questionnaire (FERUS26) for the measurement of resources and selfmanagement skills was used [29]. The FERUS26 includes 26 items to be answered on a fivepoint Likert scale. The subscales self-verbalization, coping, self-efficacy and hope can be combined to the scale "self-management skills". The internal consistency (Cronbach's alpha) for the subscales ranged between 0.86 -0.93. Patient-Provider Relationship The patients' perception of the patient-provider relationship was measured by the Patient Reaction Assessment (PRA) [30, 31]. The three subscales include communication, information and affectivity, which were measured on a seven-point Likert scale (higher scores indicate higher quality of the patient-provider relationship). The subscale information describes the perception of the patient to see the doctors as transmitters of information, who provide a more detailed description of the diseases and treatment. Communication 5
ACCEPTED MANUSCRIPT subscale assesses the ability to actively participate in the communication process. The affective subscale measures a respectful interaction of the physician with the patient and if the physician shows emotional support and understanding. The internal consistency is high (between .87 and .91). A total sum score can be calculated for determining the quality of physician-patient relationship. Cronbach´s alpha for the total score ranged between .83 -.91 in former studies [30, 31]. Statistical Analysis
AC
CE
PT E
D
MA
NU
SC
RI
PT
Assessment of the potential mediation effect of the patient-provider medical relationship relationship was evaluated using the Simple and Multiple Mediation Procedure for SPSS by Preacher and Hayes [32]. The hypothesized mediation model guiding this analysis is illustrated in Figure 1. To begin with, the criteria for a possible mediation can be tested. I) the association of the primary independent variable (attachment dimension: avoidance or anxiety) with the potential mediator (physician-patient relationship) must be significant (i.e. a ≠ 0 in the figure); furthermore II) the association of the primary independent variable (attachment dimension: avoidance or anxiety) with the dependent variable (selfmanagement) must be significant (i.e. c ≠ 0); and III) the mediator (physician-patient relationship) must be significantly associated with the dependent variable controlling for the primary independent variable (i.e. b ≠ 0). If these conditions are met and the association between the independent variable and the dependent variable is significantly reduced with the inclusion of the potential mediator in the model (i.e. c’ is statistically smaller than c) a significant mediation is assumed. The significance of the mediation effect was tested using the Simple Mediation Procedure by Preacher and Hayes [32]. The influence of the relevant sociodemographic variables and health status were controlled. An alpha level of p ≤ 0.05 was used for tests of statistical significance. Statistical analysis was performed using IBM SPSS 23.0 for Windows (Chicago, IL, USA). Furthermore, multiple mediation analyses also according to Preacher and Hayes [32] were calculated to examine whether the subscales of the PRA-D (communication, interaction and affectivity) also mediated the relationship between the attachment dimensions (anxiety and avoidance) and self-management skills. Although the number of missing values was very low, we used procedures for the management of missing values [33]. Multiple imputation procedures are commonly seen as the most adequate method for dealing with missing values in complex data sets [34]. SPSS uses a Markov-Chain-Monte-Carlo algorithm known as fully conditional specification (FCS) or chained equations imputation [35]. Multiple imputation was used to generate 10 sets of ‘‘complete’’ data with no missing values.
Figure 1 Conceptual model: the mediation effect of patient-provider relationship on attachment and self-management in older patients with multimorbidity 6
SC
RI
PT
ACCEPTED MANUSCRIPT
RESULTS
MA
NU
The socio-demographic data of the study participants are described in Table 1. A total of 219 patients (95 females) were included in the study. Patients' ages ranged from 50 to 85, with a mean age of 66.4 years ± 8.3 (Table 1). The self-reported health status ranged from 0 to 100, with a mean of 6.4 (SD = 2.1). Table 1. Characteristics of the apricare-sample (N=219)
PT E
AC
CE
Marital status
Education
Categories 50-61 62-73 74-85 Female Male Married Single Divorced Widowed Middle school Secondary modern school High school Missing Min Max
D
Variables Age M = 66.4 SD = 8.3 Sex
Number of Chronic Diseases (Physician) Multimorbidity Index (CIRS-G) Adult Attachment/ Avoidance
Frequency 67 110 42 95 124 158 9 18 34 63 112 37 7
Percentage 30.6 50.2 19.2 43.4 56.6 72.1 4.1 8.2 15.6 28.9 51.4 17.0 2.7 Mean
SD
3
18
6.4
2.5
2
38
8.3
4.2
1
7
1.9
1.3
7
ACCEPTED MANUSCRIPT
6.2
2.3
1.4
26
85
62.6
11.8
45
105
90.4
12.6
5 5 12
35 35 35
26.6 31.4 32.4
8.5 4.6 4.0
PT
1
RI
Adult Attachment/ Anxiety Self-Management and Resources (total) Patient-Provider Relationship (PRA-D total) PRA-D Subscales Communication Information Affectivity
PT E
D
MA
NU
SC
Using the Simple Mediation Procedure for SPSS by Preacher and Hayes (2004) we found that the relationship between the predictor (attachment related anxiety) and the criterion (selfmanagement skills) can be mediated through the quality of the patient-provider relationship) (Figure 2). All direct paths showed significant correlations. The path coefficient for the indirect effect remains statistically significant, therefore a partial mediation can be assumed. The confidence interval for the indirect effect ranged from -1.55 to -0.40. The overall model is considered significant Z = -3.33, p <0.001. For the model with the predictor attachment-related avoidance, the criterion self-management skills and the mediator patient-provider relationship, we found the relationships reflected in Figure 3. The confidence interval for the indirect effect ranges from -1.04 to -0.13, the path coefficient for the indirect effect is Bc' = -0.96; SEc' = 0.59 (ns), so that there is a complete mediation. The overall model is statistically significant (Z = -2.52; p <0.05).
AC
CE
Figure 2 The mediation effect of patient-provider relationship on anxiety and selfmanagement
Figure 3 The mediation effect of patient-provider relationship on anxiety and selfmanagement
8
ACCEPTED MANUSCRIPT
MA
NU
SC
RI
PT
Multiple Mediation analysis [32] considering the subscales of the PRA-D suggesting that the relationship between anxiety and self-management skills is mediated through the PRA-D subscales information and communication. The standardized regression coefficient for the indirect effect with simultaneous testing of all three subscales of the PRA-D is βc = -1.4 and is statistically significant (see Figure 4). Accordingly a partial mediation can be assumed. The overall model reached significance R2 = 0.16, p <0.001. Control variables were age, gender and the degree of chronicity (CIRS-G). In the model of avoidance and self-management only communication could be identified as a mediator. The standardized regression coefficient for the indirect effect is not statistically significant (βc´ = -0.95). We assumed full mediation (see Figure 5). The overall model is significant, R2 = 0.15, p <0.001.
AC
CE
PT E
D
Figure 4 The multivariate mediation effect of patient-provider relationship on anxiety and self-management
Figure 5 The multivariate mediation effect of patient-provider relationship on avoidance and self-management 9
RI
PT
ACCEPTED MANUSCRIPT
SC
DISCUSSION
AC
CE
PT E
D
MA
NU
The results of the mediation analyses supported the hypothesis that the patient-provider relationship mediates the relation between adult attachment and self-management in patients with multi-morbidity. We were also able to show that the relation between attachment-related anxiety and self-management was mediated by the subscales information and communication of the PRA-D and the relation between attachment-related avoidance and self-management was mediated by the subscale communication as parts of the patient-provider medical relationship. The communication scale reflects the patients’ experience in initiating communication with the GP concerning some aspect of their illness and treatment. It was confirmed that communication is one of the most important aspects in the patient-provider relationship [36, 37]. Therefore, it is not surprising that communication (such as the ability to ask questions during treatment) also has an influence on the relationship between patient characteristics and self-management in anxiety and in avoidant patients. This is consistent with the results of Ciechanowksi et al. (2001) who also identified the subscale communication as an influential factor in diabetes patients with avoidant attachment on health-related outcomes [38]. However, the information transfer to patients is an important task for physicians, and is the basis for behavioral change in selfmanagement [39]. In our study, the subscale information measured if patients had been given information about the possible side effects of their treatment/medication, and whether the treatment plan had been presented to them transparently for the following weeks and months. This could have an influence on the self-management skills of patients with anxious attachment, but only long-term studies or trials can answer this more precisely. Patients with anxious attachment do not feel resilient and resistant by the chronic activation of the attachment system [40]. They are often prone to vigorously searching the proximity of the other [41] , catastrophizing in representation of symptoms [42] and exaggerating their expression of needs to get support from their caregiver [43].Particularly patients with anxious attachment need clear, concise information about the course of their treatment 10
ACCEPTED MANUSCRIPT
RI
PT
this can enhance their self-management skills. In contrast, avoidant patients are characterized by high levels of self-reliance, greater interpersonal distance [14], as well as minimal help seeking behavior and expression of distress [15, 16]. They cope with stress with cognitive distancing from their emotions, denial and distraction [17], and seek information about their disease more independently of the GP as they tend to distrust others, expect them to be hostile, exploitative and not sensitive [44, 45]. In contrast, the subscale affectivity could not be identified as a mediator between the attachment-related anxiety and avoidance and self-management. This might be due to methodological limits of this subscale (e.g. a low internal consistency, alpha = 0.70, cf. [31]). On the other hand, the construct of emotional support cannot clearly be separated from the constructs of communication and information in the context of the doctor-patient-provider relationship because parts of it are already included.
MA
NU
SC
Patient surveys from a European Community study "European Project on Patient Evaluation of General Practice Care" (EUROPEP) have published the five most desirable qualities they wish for in a GP, i.e. willingness to answer questions clearly and concisely and to listen, quick accessibility, and confidentiality of patient data, information, and sufficient consultation time [46]. None of the requests explicitly or exclusively mention the affective components of the patient-provider relationship. Nevertheless, we think it is an important factor, but future research should examine the influence of the affective component (respectful interactions with the patient, emotional support and understanding) in more detail.
AC
CE
PT E
D
It has already been proven, that the individual attachment style of the patients can influence their self-care skills and behavior [9-11]. Additionally, we were able to find indications in this study that the quality of the patient-provider relationship can mediate this association. In order to improve the self-management in patients with anxiety attachment, the practitioner should give more structured information about methods of self-treatment and the side effects of medication. For these patients, the GP often is the main point of contact. In contrast, patients with avoidant attachment seek information about treatment, selftreatment and side effects more externally in books or on the internet [47]. If physicians reflect the effects of different attachment styles, they can provide positive and beneficial relationships and communication styles in different patients with chronic conditions [13, 44].
11
ACCEPTED MANUSCRIPT REFERENCES
AC
CE
PT E
D
MA
NU
SC
RI
PT
1. E.H. Wagner, B.T. Austin, C. Davis, M. Hindmarsh, J. Schaefer, A. Bonomi, Improving chronic illness care: translating evidence into action, Health Aff (Millwood) 20 (2001) 64-78. 2. WONCA Europe Council, The European definition of general practice / family medicine. http://www.woncaeurope.org/sites/default/files/documents/Definition%203rd%20ed%202 011%20with%20revised%20wonca%20tree.pdf, 2002 (accessed 12.12.16). 3. M. Huber, J.A. Knottnerus, L. Green, H. van der Horst, A.R. Jadad, D. Kromhout, B. Leonard, K. Lorig, M.I. Loureiro, J.W. van der Meer, P. Schnabel, R. Smith, C. van Weel, H. Smid, How should we define health?, BMJ 343 (2011) d4163, doi: 10.1136/bmj.d4163. 4. J. Bowlby, A Secure Base. Clinical Applications of Attachment Theory, Routledge, London, 1988. 5. W.H. Berman, L. Marcus, E.R. Berman, Attachment in marital relations, in: M.B. Sperling, W.H. Berman (Eds.), Attachment in adults: theory, assessment and treatment. Guilford Press, New York, 1994. 6. P.R. Shaver, M. Mikulincer, What do self-report attachment measures assess?, in: W.S. Rholes, J.A. Simpson (Eds.), Adult attachment: theory, research, and clinical implications. Guilford Press, New York, 2004, pp. 17-54. 7. B. Strauss, K. Brenk-Franz, The relevance of attachment theory in medical care, in: J.J. Hunter, R.G. Maunder (Eds.), Improving patient treatment with attachment theory: A guide for primary care practitioners and specialists. Springer, 2016. 8. R.G. Maunder, J.J. Hunter, Attachment and psychosomatic medicine: developmental contributions to stress and disease, Psychosom Med 63 (2001) 556-567. 9. P. Ciechanowski, J. Russo, W. Katon, M. Von Korff, E. Ludman, E. Lin, G. Simon, T. Bush, Influence of patient attachment style on self-care and outcomes in diabetes, Psychosom Med 66 (2004) 720-728, doi: 10.1097/01.psy.0000138125.59122.23. 10. P.S. Ciechanowski, J.E. Russo, W.J. Katon, E.A. Walker, Attachment theory in health care: the influence of relationship style on medical students' specialty choice, Med Educ 38 (2004) 262-270, doi: 1767 [pii]. 11. K. Brenk-Franz, B. Strauss, F. Tiesler, C. Fleischhauer, P. Ciechanowski, N. Schneider, J. Gensichen, The influence of adult attachment on patient self-management in primary care the need for a personalized approach and patient-centred care, PLoS One 10 (2015) e0136723, doi: 10.1371/journal.pone.0136723. 12. R.C. Fraley, N.G. Waller, K.A. Brennan, An item response theory analysis of self-report measures of adult attachment, J Pers Soc Psychol 78 (2000) 350-365. 13. D. Thompson, P.S. Ciechanowski, Attaching a new understanding to the patientphysician relationship in family practice, J Am Board Fam Pract 16 (2003) 219-226. 14. J.G. Allen, H. Stein, P. Fonagy, J. Fultz, M. Target, Rethinking adult attachment: a study of expert consensus, Bull Menninger Clin 69 (2005) 59-80, doi: 10.1521/bumc.69.1.59.62266. 15. W.S. Rholes, J.A. Simpson, J.G. Stevens, Attachment orientations, social support, and conflict resolution in close relationships, in: J.A. Simpson, W.S. Rholes (Eds.), Attachment theory and close relationships. Guilford Press, New York, 1998, pp. 166-188.
12
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
16. M. Main, Avoidance in the service of attachment: a working paper, in: K. Immelman, G. Barlow, M. Main, L. Petrinovitch (Eds.), Behavioral development: the Bielefeld Interdisciplinary Project. Cambridge University Press, New York, 1981, pp. 651–693. 17. J.A. Feeney, Adult attachment, coping style and health locus of control as predictors of health behavior, Aust J Psychol 47 (1995) 171-177. 18. J.J. Hunter, R.G. Maunder, Using attachment theory to understand illness behavior, Gen Hosp Psychiatry 23 (2001) 177-182, doi: S0163-8343(01)00141-4 [pii]. 19. H.B. Frederiksen, J. Kragstrup, B. Dehlholm-Lambertsen, Attachment in the doctorpatient relationship in general practice: a qualitative study, Scand J Prim Health Care 28 (2010) 185-190, doi: 10.3109/02813432.2010.505447. 20. G. D'Elia, [Attachment theory: a biological basis for the patient-physician relation?], Lakartidningen 97 (2000) 1956-1959. 21. L. Simmons, N.J. Baker, J. Schaefer, D. Miller, S. Anders, Activation of patients for successful self-management, J Ambul Care Manage 32 (2009) 16-23, doi: 10.1097/01.JAC.0000343120.07844.a9. 22. K.H. Jöckel, B. Babitsch, B.M. Bellach, K. Bloomfield, J.H.P. Hoffmeyer-Zlontnik, J. Winkler, C. Wolf, [Recommendations of the working group "Epidemiologic Methods" in the German Association Epidemiology of GMDS and DGSMP to measure and quantify the sociodemographic characteristics in epidemiological studies], in: W. Ahrens, B.M. Bellach, K.H. Jöckel (Eds.), [Measurement of socio-demographic characteristics in epidemiology]. MMVVerlag, Munich, 1998, pp. 7-38. 23. The EuroQol Group, Euro-Qol - a new facility for the measurement of health-related quality of live, Health Policy 16 (1990) 199-208. 24. J.-M. Graf von der Schulenburg, C. Claes, W. Greiner, A. Uber, [The German version of the EuroQol questionnaire], Z f Gesundheitswiss 6 (1998) 3-20. 25. M.D. Miller, C.F. Paradis, P.R. Houck, S. Mazumdar, J.A. Stack, A.H. Rifai, B. Mulsant, C.F. Reynolds, 3rd, Rating chronic medical illness burden in geropsychiatric practice and research: application of the Cumulative Illness Rating Scale, Psychiatry Res 41 (1992) 237248, doi: 0165-1781(92)90005-N [pii]. 26. C. Hudon, M. Fortin, H. Soubhi, Abbreviated guidelines for scoring the Cumulative Illness Rating Scale (CIRS) in family practice, J Clin Epidemiol 60 (2007) 212, doi: 10.1016/j.jclinepi.2005.12.021. 27. J.C. Ehrenthal, U. Dinger, A. Lamla, B. Funken, H. Schauenburg, [Evaluation of the German version of the attachment questionnaire "Experiences in Close Relationships-Revised" (ECR-RD)], Psychother Psychosom Med Psychol 59 (2009) 215-223, doi: 10.1055/s2008-1067425. 28. K. Brenk-Franz, J. Ehrenthal, T. Freund, N. Schneider, B. Strauss, F. Tiesler, H. Schauenburg, J. Gensichen, Evaluation of the Short Form of “Experience in Close Relationships” (Revised, German Version “ECR-RD12”) - a Tool to Measure Adult Attachment in Primary Care PLoS One (submitted). 29. M. Jack, FERUS - Questionnaire for Measuring Resources and Self-management skills, Hogrefe, Göttingen, 2007. 30. J.P. Galassi, R. Schanberg, W.B. Ware, The Patient Reactions Assessment: a brief measure of the quality of the patient-provider medical relationship, Psychol Assess 4 (1992) 346-351. 13
ACCEPTED MANUSCRIPT
AC
CE
PT E
D
MA
NU
SC
RI
PT
31. K. Brenk-Franz, G. Hunold, J.P. Galassi, F. Tiesler, W. Herrmann, T. Freund, C. SteurerStey, S. Djalali, A. Sönnichsen, N. Schneider, J. Gensichen, [Quality of the physician-patient relationship – evaluation of the German version of the Patient Reactions Assessment (PRAD)], Z Allg Med 92 (2016) 103-108. 32. K.J. Preacher, A.F. Hayes, Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models, Behav Res Methods 40 (2008) 879891. 33. J.L. Schafer, Analysis of incomplete multivariate data, London, Chapman & Hall, 1997. 34. D.B. Rubin, Multiple imputation after 18+ years (with discussion), JASA (1996) 473489. 35. W.R. Gilks, S. Richardson, D.J. Spiegelhalter, Markov Chain Monte Carlo in Practice, Chapman and Hall, London, 1996. 36. L.M. Ong, J.C. de Haes, A.M. Hoos, F.B. Lammes, Doctor-patient communication: a review of the literature, Soc Sci Med 40 (1995) 903-918. 37. K. Laemmel, [Interview in medicine], Praxis (Bern 1994) 85 (1996) 863-869. 38. P.S. Ciechanowski, W.J. Katon, J.E. Russo, E.A. Walker, The patient-provider relationship: attachment theory and adherence to treatment in diabetes, Am J Psychiatry 158 (2001) 29-35. 39. M. Lipkin, Jr., Physician-patient interaction in reproductive counseling, Obstet Gynecol 88 (1996) 31S-40S. 40. P.J. Meredith, J. Strong, J.A. Feeney, Evidence of a relationship between adult attachment variables and appraisals of chronic pain, Pain Res Manag 10 (2005) 191-200. 41. M. Mikulincer, P.R. Shaver, Attachment in adulthood: structure, dynamics, and change, Guilford, New York, 2007. 42. P. Ciechanowski, M. Sullivan, M. Jensen, J. Romano, H. Summers, The relationship of attachment style to depression, catastrophizing and health care utilization in patients with chronic pain, Pain 104 (2003) 627-637, doi: S0304395903001209 [pii]. 43. A. Slade, Attachment theory and research: implications for the theory and practice of individual psychotherapy with adults, in: J. Cassidy, P.R. Shavers (Eds.), Handbook of attachment: theory, research, and clinical applications. Guilford, New York, 1999. 44. P. Ciechanowski, W.J. Katon, The interpersonal experience of health care through the eyes of patients with diabetes, Soc Sci Med 63 (2006) 3067-3079, doi: 10.1016/j.socscimed.2006.08.002. 45. M. Dozier, K.L. Cue, L. Barnett, Clinicians as caregivers: role of attachment organization in treatment, J Consult Clin Psychol 62 (1994) 793-800. 46. A. Klingenberg, O. Bahrs, J. Szecsenyi, [How do patients evaluate general practice? German results from the European Project on Patient Evaluation of General Practice Care (EUROPEP)], Z Arztl Fortbild Qualitatssich 93 (1999) 437-445. 47. J.J. Hunter, R.G. Maunder, Foundations. Introduction, in: J.J. Hunter, R.G. Maunder (Eds.), Improving patient treatment with attachment theoy: A guide for primary care practitioners and specialists. Springer, 2016.
14
ACCEPTED MANUSCRIPT
Highlights
Adult attachment is associated with self-management of patients with chronic conditions. The quality of the patient-provider relationship mediated the link between adult attachment and self-management among primary care patients with multiple chronic diseases.
PT
Particularly the quality of communication and information over the course of treatment has a significant mediating influence.
AC
CE
PT E
D
MA
NU
SC
RI
An active patient-provider communication and gives information about the treatment to the patient may improve self-management skills in patients.
15