Professional Documents
Culture Documents
The Development and Psychometric Validation of The Self-Efficacy and Performance in Self-Management Support (SEPSS) Instrument
The Development and Psychometric Validation of The Self-Efficacy and Performance in Self-Management Support (SEPSS) Instrument
The Development and Psychometric Validation of The Self-Efficacy and Performance in Self-Management Support (SEPSS) Instrument
Correspondence to V. Duprez: DUPREZ V., VAN HOOFT S.M., DWARSWAARD J., VAN STAA A.L., VAN
e-mail: veerle.duprez@ugent.be H E C K E A . & S T R A T I N G M . M . H . ( 2 0 1 6 ) The development and psychometric val-
idation of the self-efficacy and performance in self-management support (SEPSS)
*Both authors contributed equally to this
Instrument. Journal of Advanced Nursing 72(6), 13811395. doi: 10.1111/
work.
jan.12918
Veerle Duprez MSc RN
PhD Candidate Abstract
University Centre for Nursing and Aim. To develop and psychometrically test the self-efficacy and performance in
Midwifery, Department of Public Health, self-management support (SEPSS) instrument.
Faculty of Medicine and Health Sciences, Background. Facilitating persons with a chronic condition to take an active role
Ghent University, Ghent, Belgium
in the management of their condition, implicates that nurses acquire new
@ucvvgent.be
competencies. An instrument that can validly and reliably measure nurses
Susanne M. Van Hooft MSc RN performance and their perceived capacity to perform self-management support is
PhD Candidate needed to evaluate current practice and training in self-management support.
Research Centre Innovations in Care, Design. Instrument development and psychometric testing of the content and
Rotterdam University of Applied Sciences, construct validity, factor structure and reliability.
the Netherlands and Institute of Health Methods. A literature review and expert consultation (N = 17) identified the
Policy & Management, Erasmus University content. The items were structured according to the Five-As model and an
Rotterdam, Rotterdam, the Netherlands
overarching category of overall competencies. The initial instrument was tested
@hsrotterdam
in a sample of 472 nurses and 51 nursing students from Belgium and the
Jolanda Dwarswaard PhD Netherlands, between June 2014January 2015.
Senior Researcher Results. Confirmatory factor analyses revealed satisfactory fit indices for the six-
Research Centre Innovations in Care, factor structure. Discriminating power was demonstrated for subgroups. The
Rotterdam University of Applied Sciences, overall internal consistency (Cronbachs alpha) was high both for the self-efficacy
the Netherlands and Institute of Health and the performance items. The testretest intra-class correlation coefficients were
Policy & Management, Erasmus University good.
Rotterdam, Rotterdam, the Netherlands
Conclusion. The SEPSS instrument is a 36-item, Likert-scaled self-reporting
instrument with good content and construct validity, and good internal
AnneLoes van Staa MD PhD RN
Professor Transitions in Care consistency reliability and good testretest reliability. Therefore, it is a promising
Research Centre Innovations in Care, instrument to measure self-efficacy and performance with regard to
Rotterdam University of Applied Sciences, self-management support.
the Netherlands and Institute of Health
Policy & Management, Erasmus University Keywords: competencies, instrument development, nursing, psychometric, relia-
Rotterdam, Rotterdam, the Netherlands
bility, self-management support, validity
continued on page 1382
(Schulman-Green et al. 2012, Dwarswaard et al. 2015). self-management and patient centeredness, which all
The Arrange phase refers to organizing follow-up care. Self- showed good internal consistency. However, nursing com-
management support is a multidisciplinary approach which petencies to stimulate patients to take the lead in their self-
relies on effective information sharing and effective coordi- management are not addressed in detail.
nation of care (Pols 2009). Importantly, arrangements must These existing instruments typically focus on performance
be made to evaluate the progress in goal achievement (Glas- in SMS. It may be the case, however, that healthcare profes-
gow et al. 2003). In addition, nurses need to possess overall sionals have the required skills, but lack self-efficacy to effec-
competencies for a partnership attitude in each phase of the tively apply these skills (Bandura 1991, Kosmala-Anderson
support process. This includes respecting patients auton- et al. 2010a,b). Self-efficacy refers to a persons confidence in
omy in shared decision-making, building a sustainable part- the ability to perform a specific behaviour in a specific situa-
nership and being able to reflect on ones own actions and tion (Bandura 1991). Self-efficacy is known to affect beha-
recognize ethical dilemmas (Hostick & McClelland 2002, viour by influencing the choices individuals make and the
Pols 2009, Sandman et al. 2012, Kayser et al. 2014). course of actions they pursue; it determines their level of
Studies reveal a discrepancy between the expected profi- effort, persistence and resilience (Bandura 2006).
ciency of nurses and their actual performance on SMS (Elis- The current evidence demonstrates that other factors than
sen et al. 2013, Yank et al. 2013). One of the ways to self-efficacy might affect a nurses performance in SMS (Har-
improve the provision of SMS in chronic care is the training ris et al. 2008, Elissen et al. 2013), creating the potential risk
of healthcare providers (Zwar et al. 2006, Kosmala-Ander- of a discrepancy between self-efficacy and performance.
son et al. 2010a,b). Training is also likely to improve self- Therefore, it is appropriate to develop an instrument that
efficacy and thus performance of SMS as self-efficacy is a measures not only nurses actual performance but also self-
strong predictor of behaviour (Bandura 1991) and thereby efficacy to perform SMS for people with chronic conditions.
an important precursor of SMS performance. To the best of
our knowledge, there is no instrument to evaluate the confi-
The study
dence nurses have in their own SMS abilities.
A valid and reliable instrument assessing both perfor-
Aim
mance and self-efficacy is useful to guide and measure the
current practice, to identify educational needs and to assess To develop and psychometrically test the Self-efficacy and Per-
the effectiveness of training programs. formance in Self-Management Support (SEPSS) instrument.
Background Methodology
Several instruments are available to measure healthcare pro- A psychometric instrument validation study was conducted
fessionals performance in SMS. These only address specific in two phases. Phase one included instrument development
aspects, however. The Clinician Support-Patient Activation and the process of content validation by a panel of experts.
Measure (CS-PAM) measures beliefs about the importance Phase two entailed the psychometric evaluation in a sample
of activating patients and of SMS (Hibbard et al. 2010). of nurses and nursing students (Figure 1).
Decision support can be addressed with instruments such as
the Observing Patient Involvement (OPTION) scale (Elwyn Phase 1 Instrument development & content validation
et al. 2013), the Shared Decision Making Questionnaire First, a literature and concept search in scientific and grey
physician version (SDM-Q-Doc) (Scholl et al. 2012) and literature was performed (MarchNovember 2013) to iden-
the Decision Support Analysis Tool (DSAT-10) (Stacey tify relevant competencies for SMS. We searched in the
et al. 2008). Therapeutic alliance can be measured with the PubMed, CINAHL and Cochrane databases for scientific
Kim Alliance Scale (KAS) (Kim et al. 2001); and skills in articles about the concept of self-management and the
motivational interviewing with, for example, the Motiva- required competencies for SMS, using the keywords self-
tional Interviewing Treatment Integrity (MITI) (Moyers care, chronic disease, nurs* and competenc*. We also
et al. 2005) or the Behavior Change Counselling Scale retrieved information from (inter)national policy documents
(BCCS) (Vallis 2013). To our knowledge, only the Practices on self-management. The processes of self-management in
in SMS (PSMS) covers the broad aspect of SMS (Kosmala- patients with chronic conditions, consisting of patient tasks
Anderson et al. 2011). This 25-item instrument has three and skills as described by Schulman-Green et al. (2012),
subscales: clinician SMS, organization of services to support formed the basis for a draft list. These processes were con-
verted into competencies for SMS. Additionally, competen- assess both self-efficacy and performance in SMS, each item
cies such as partnership (Hostick & McClelland 2000, Lei- was assessed by two questions (additional File S1). Self-effi-
sen & Hyman 2001, Keatinge et al. 2002, Lorig & cacy was measured by requesting participants to consider I
Holman 2003, Visse et al. 2010), shared decision-making, think I can do this, with ratings from Not at all(0), Not
collaborative goal setting (Lorig & Holman 2003, Stacey sufficient(1), More or less(2), Sufficient(3), Good(4).
et al. 2008, Kriston et al. 2010) and self-efficacy of the Actual performance was measured by requesting partici-
patient (Krichbaum et al. 2003, Lorig & Holman 2003, pants to consider I do this, with ratings from Never(0),
Yank et al. 2013) were obtained from literature. The items Rarely(1), Occasionally(2), Frequently(3)-Always(4).
in the list were structured according to the Five As model
described above (Glasgow et al. 2003). An overarching sixth Phase 2 Psychometric evaluation
category was added to cover overall competencies for SMS The psychometric evaluation (Figure 1) included the testing
that could not be related to one single step of the Five As of the construct validity (confirmatory factor analysis, dis-
model (Leisen & Hyman 2001, Hostick & McClelland 2002, criminating power) and reliability (internal consistency and
Glasgow et al. 2003, Pols 2009, Kriston et al. 2010, Visse stability) of the SEPSS instrument.
et al. 2010). In the end, the draft list contained 37 competen-
cies, grouped into six subscales: (1) Assess assess the needs Sample
and beliefs of the patient, (2) Advise give the patient infor- The 46-item instrument was tested in a sample of nurses
mation he needs, (3) Agree set goals together with the and nursing students in Belgium and the Netherlands. The
patient, (4) Assist assist the patient to overcome barriers, sample size aimed for was based on the recommended 10
(5) Arrange arrange follow-up care and (6) Overall compe- respondents per item as a minimum to support the factor
tencies - a supportive attitude (Table 2). analysis for stable covariates (Polit & Beck 2008). A total
This draft list was discussed by a convenience sample of sample approach was used. In Belgium, 122 final-year nurs-
experts in SMS (N = 10) during a 3-hour meeting. Given that ing students were invited (response 51/122; 42%) and 58
the instrument should be appropriate for all healthcare set- nurses combining their employment with attending an addi-
tings and for educational purposes, the experts represented tional Master of Science in Nursing program (response 37/
nurse education, hospital care, older people care and psychi- 58; 64%) participated. In the Netherlands, we invited 2054
atric care. During the meeting the relevance, appropriateness nurses from an academic hospital and 107 nurses from a
and exhaustiveness of the item pool were discussed. Following psychiatric institution. Respectively 345 (17%) and 32
on from the qualitative comments of the experts, three compe- (30%) participated in the validation study. Furthermore,
tencies were excluded, three competencies were reformulated 800 nurses employed in different healthcare settings and
and six competencies were added. This resulted in a 40-item participating in a Dutch national panel of nurse profession-
draft instrument. The grouping into the six subscales was als were invited (response 58/800; 7%). This resulted in a
approved by the experts. In the next step, the researchers split total of 523 participants.
broad competencies into sub-competencies to allow detailed
assessment, which increased the number of items to 53. Procedure
The relevance and clarity of the 53-item instrument were Data were collected between June 2014 January 2015.
pilot-tested in a new group of experts in SMS (N = 4), The nursing students completed a paper form of the self-
nurses (N = 8) and researchers (N = 5). This resulted in reporting instrument. The nurses completed the question-
some minor adjustments that entailed mainly wording naire in an online format. Next to the items of the SEPSS,
ambiguities and in a reduction by seven items due to over- participants were asked for demographic variables and their
lap in content or meaning. To cover the content of each perception of the importance of SMS, on a scale ranging
subscale and to allow for items to be deleted during the from 1 (not important at all)-10 (very important). To
psychometric testing and refinement of the instrument, at increase the response rate, for the online procedure, two
least six items were included for each subscale. Phase one reminders were sent and small rewards (e.g. movie tickets)
resulted in an initial 46-item instrument with established were raffled among the participants. As the instrument can
content validity, grouped into 6 subscales (Figure 1). be used to measure current practice in SMS, its stability
was evaluated using the testretest procedure. For this pur-
Instrument pose, a group of nursing students (N = 26) completed the
The items were formulated to be measured on a five-point instrument twice, with a 2-hour interval. This short interval
Likert rating scale. As the aim of the instrument was to was chosen to minimize the possible effect of confounding
Reliability
Test-retest reliability (ICC)
..
SEPPS Instrument (36 items)
Self-efficacy: Cronbachs = 096; ICC = 095 (95% Cl = 088098)
Performance: Cronbachs = 095; ICC = 094 (95% Cl = 085098)
SEPSS Subscales
1. Assess (6 items)
a. Self-efficacy: Cronbachs = 085; ICC = 092 (95% Cl = 080096)
b. Performance: Cronbachs = 084; ICC = 085 (95% Cl = 067094)
2. Advise (6 items)
a. Self-efficacy: Cronbachs = 082; ICC = 095 (95% Cl = 087098)
b. Performance: Cronbachs = 075; ICC = 096 (95% Cl = 090098)
3. Agree (6 items)
a. Self-efficacy: Cronbachs = 089; ICC = 090 (95% Cl = 078096)
b. Performance: Cronbachs = 088; ICC = 093 (95% Cl = 083097)
4. Assist (6 items)
a. Self-efficacy: Cronbachs = 087; ICC = 084 (95% Cl = 062094)
b. Performance: Cronbachs = 085; ICC = 086 (95% Cl = 074095)
5. Arrange (6 items)
a. Self-efficacy: Cronbachs = 084; ICC = 091 (95% Cl = 079096)
b. Performance: Cronbachs = 082; ICC = 086 (95% Cl = 068094)
6. Overall competencies (6 items)
a. Self-efficacy: Cronbachs = 083; ICC = 087 (95% Cl = 069094)
b. Performance: Cronbachs = 081; ICC = 083 (95% Cl = 060093)
factors, such as learning by lectures or experiences on clini- about the testretest procedure, making the procedure less
cal placement and by spontaneous growth (Polit & Beck sensitive to memory bias. The conditions were the same for
2008). The participants were not informed in advance both parts of the procedure.
effects were present if >15% of the nurses achieved values Table 1 Demographic characteristics of the sample.
in the 125% lower and upper bound, respectively, of (sub)
Characteristics (N = 523) N (%)
scale values.
Gender
Female 409 (782)
Translation Male 110 (210)
Missing 4 (08)
For international publication and presentation purposes the Age (years)
initial 46-item instrument was translated from Dutch into <23 43 (82)
English by an independent native speaker. Another indepen- 2329 144 (275)
dent professional translator re-translated the items in 3039 104 (199)
4049 96 (184)
Dutch. The re-translated version was compared with the
>49 132 (252)
original wordings, to confirm the accuracy of the English Missing 4 (08)
translation. Discrepancies between the translations were Setting
resolved by consensus. Student nurses 51 (97)
Academic hospital
Inpatient units 269 (514)
Ethical considerations Outpatient clinics 87 (166)
General hospital
In Belgium, the study protocol was approved by the Inpatient units 24 (46)
Ethical Review Committee of Ghent University Hospital Psychiatric institution 33 (64)
(B670201422154 and B670201422381). While in the Primary & elderly care nursing 9 (17)
Netherlands no Research Ethics Committee approval Other (not specified) 50 (96)
Work experience (years)
was required, permission was obtained from the execu-
05 124 (237)
tive boards of all participating institutions. All partici- 610 97 (185)
pants received detailed information about the aim and 1115 58 (111)
procedures and were informed of confidentiality. The >15 171 (327)
nursing students gave their written informed consent Missing 73 (140)
Educational degree
before completing the instrument. For the other partici-
Student nurses, vocational educational level 51 (97)
pants, completing the online survey was considered as
Vocational education level* 100 (191)
consent. Bachelor degree 268 (513)
Master degree 59 (113)
Missing 45 (86)
Results
*Vocational educational level is a three years nurse training educa-
tion at qualification level 5 of the European Higher Education
Sample characteristics
Area.
The sample included 472 nurses and 51 nursing students. Both academic and professional Master degrees.
The nurses worked in different settings, more than half
of them (56%) on inpatient units in a general or aca- tor loadings, modification indices and an internal consis-
demic hospital. About one sixth of the nurses (166%) tency check of each subscale, the stepwise procedure, as
worked on an outpatient clinic providing consultations described in the method section, resulted in the elimination
with chronically ill on a daily basis. For further details of 10 items (bold in Table 2). The final model consisted of
see Table 1. 36 items with six items for each subscale. This final model
resulted in a better fit of the model, although the fit indices
still showed room for improvement; v2 decreased to 7238;
Construct validity
RMSEA decreased to 012; SRMR decreased to 010 and
Factor analysis IFI increased to 093. A similar procedure was done for the
The confirmatory factor analysis on the self-efficacy items performance items, resulting in a similar fit of the model
yielded the following results: v2 was 12086; RMSEA 013; for both the initial and the final model. Also, the exact
SRMR 011 and IFI 090 all indicating that the model was same items were removed following the procedure for
not yet sufficient. Factor loadings of this initial 46-item improving the model. Sample adequacy was confirmed by
model ranged from 044-087 (Table 2). Following the fac- the KMO test (095) and Bartletts test of sphericity
1388
Self-efficacy Performance
Subscale Assess
1. Asking the patient what he expects from living with a (chronic) condition in the near 520 289 086 520 181 103 073
future
2. Asking the patient about his own experiences with his (chronic) condition 520 315 077 519 231 099 063
3. Asking the patient what he knows about his (chronic) condition 520 316 075 520 231 106 075
4. Asking the patient about how he can share his emotions about the (chronic) condition 521 300 083 519 214 107 070
with important others
5. Asking the patient about the available motivation and discipline to integrate the chronic 521 270 092 518 172 106 072
condition in his life
6. Asking the patient how much confidence he has in his own abilities 520 282 088 517 183 101 066
7. Asking the patient what he can and will do in his daily health care 520 320 078 517 252 108 072
8. Asking the patient which fundamental values (e.g. religious, cultural, independence) are of 519 251 100 520 144 103 062
influence of his perception of the condition
Subscale Advise
9. During each contact, asking the patient what information he needs 484 303 079 483 227 103 079
10. Asking the patient for permission before giving information or advice 483 276 092 480 168 114 069
11. Letting the patient restate the information that I gave 482 284 084 480 182 101 081
12. Giving the patient information and instruction about the (chronic) condition 481 316 080 479 250 103 067
13. Helping the patient to formulate questions to discuss with other healthcare professionals 483 270 093 480 161 101 066
14. Informing the patient of the choices he has (which he can discuss with other healthcare 482 275 090 479 178 105 055
professionals)
15. Involving the family when providing information and instruction 479 320 076 479 240 111 066
Subscale Agree
16. Helping the patient to identify earlier positive experiences with achieving goals 452 264 089 447 156 102 048
17. Allowing the patient to determine his own priorities when developing goals 451 268 086 448 156 106 074
18. Jointly with the patient, developing a plan of action to achieve the goals 452 252 098 446 134 109 055
19. Documenting the goals and agreements in the patients record 452 282 100 448 200 131 079
20. Helping the patient to make decisions concerning his treatment jointly with me and/or 451 257 094 448 153 104 056
the other healthcare professionals
21. Recognizing the patients anxiety about making a treatment decision 452 292 084 446 194 103 068
Subscale Assist
22. Inviting the patient to talk about deteriorating health and changes in his life 423 277 091 424 188 108 052
23. Discussing with the patient who he will inform about his chronic condition 423 258 097 420 134 112 067
24. Stimulating the patients self-confidence so that he can integrate the chronic condition in 426 283 085 422 195 108 061
his life
25. Encouraging the patient to perform as many daily living activities as possible 425 316 074 423 259 099 073
26. Helping the patient to choose the activities that he can realistically perform 423 298 074 421 223 101 062
27. Discussing with the patient who (i.e. family, friends, network) can provide daily support 421 300 078 420 216 108 081
28. Discussing with the patient how he can make use of self-management assistive devices 421 248 105 420 138 115 073
35. Facilitating the patient to easily stay in contact between appointments 409 286 102 404 208 139 087
36. Initiating contact between appointments with the patient, to discuss his health and to 407 244 121 405 116 123 072
solve possible difficulties
37. Together with the patient, examining progress of the care plan actions 408 251 104 405 134 116 072
Subscale Overall Competencies
38. Valuing and respecting the patient as a partner in his care 402 330 075 399 297 100 054
39. Acknowledging the patients experiential knowledge as valuable information concerning 402 328 068 399 283 092 079
my own care delivery
40. Considering the (cultural) background of the patient 401 317 070 400 287 095 066
41. Together with the patient, determining how much of the care coordination I take over 399 297 081 399 240 111 074
for him
42. Using the patients choice as the basis for care, even if it is not ideal from a medical 399 274 086 399 196 108 064
perspective
43. Showing understanding when the patient does not succeed in achieving the established 400 305 080 398 236 109 074
goals
44. Deviating from protocols when necessary 401 301 085 398 176 104 061
45. Reflecting on my own management (of care) 400 326 070 398 273 092 077
46. Applying principles of negotiation and conflict-management 400 277 087 398 193 098 058
1389
Validation of self-efficacy and performance in self-management support instrument
V. Duprez et al.
(v2 = 765423, d.f. = 630, P < 0001) indicating that corre- ceiling effect for the Overall Competence scale concerning
lations between items did not occur by chance. self-efficacy.
Discriminating power
Discussion
The results on discriminating power demonstrated signifi-
cant differences between most of the predefined groups, as As self-management has become the leading paradigm for
shown in Table 3. Nurses providing outpatient consulta- chronic care in many countries, it is essential to develop
tions had higher scores than nurses in inpatients units at all SMS training programs for nurses and to measure the effec-
subscales and at the total scale level for self-efficacy (respec- tiveness of these programs. In this regard, the SEPSS instru-
tively 1871 vs. 1675, t = 370, d.f. = 7890, P < 0001) ment provides for accurate assessment of a nurses
and for performance (respectively 1399 vs. 1147, t = 417, performance and self-efficacy in applying SMS. Other than
d.f. = 7858, P < 0001). Nurses had higher scores than the PSMS instrument (Kosmala-Anderson et al. 2011), the
nursing students at all subscales and at the total scale level SEPSS places an emphasis on competencies needed to stimu-
for self-efficacy (total scores respectively 1722 vs. 1606, late patients to take the lead in self-managing their chronic
t = 221, d.f. = 394, P < 005) and for performance (respec- condition.
tively 1202 vs. 939, t = 423, d.f. = 391, P < 0001). The SEPSS instrument assesses the performance and the
Nurses who perceived SMS as highly important had higher self-efficacy of essential competencies for SMS derived from
scores for self-efficacy than nurses believing SMS of little or literature and expert advice, complemented with competen-
no importance for chronic care, (total scores respectively cies reflecting key attitudes, such as partnership and patient
1775 vs. 1624, t = 210, d.f. = 108, P < 005) and for per- centred-care. It relies on a broad holistic perspective on
formance (total scores respectively 1260 vs. 1133, SMS, based on what patients need to take the lead in self-
t = 173, d.f. = 108, P < 005). Nurses with a master managing their chronic condition (Schulman-Green et al.
degree had higher levels of performance than those without 2012). Although the instrument uses the framework of the
such a degree (total scores respectively 1300 vs. 1154, Five As model, familiarity with this model is not a prereq-
t = 238, d.f. = 7416, P < 005), but self-efficacy did not uisite for using the SEPSS. The underlying competencies are
significantly differ between these groups (1748 vs. 1707, feasible for all professionals supporting self-management.
t = 094, d.f. = 366, P = 035). Regarding construct validity of the SEPSS, the confirma-
tory factor analysis yielded satisfactory fit with the 36-item
SEPSS instrument, wherein the overall competencies can
Reliability
be considered as overarching for the other five subscales
Internal consistency according to the Five As model. By removing 10 items, we
Cronbachs alpha was 096 for the total self-efficacy scale. aimed to develop a brief instrument that still has enough
For the subscales of self-efficacy and performance, Cron- sensitivity to measure what it is supposed to measure. For
bachs alpha values are displayed in Figure 1. that reason, we did not allow a <080 and maintained at
least six items in each subscale. Although the fit indices
Testretest stability showed room for improvement, factor loadings were high
A group of 26 final-year nursing students completed the and sample adequacy to perform the factor analysis was
questionnaire twice. On the first occasion the mean total confirmed by the KMO test and Bartletts test of sphericity.
score for self-efficacy was 1684 (SD 365) and for perfor- The results of the known-group technique analysis sup-
mance in SMS 1045 (SD 428). At retest, the corresponding ported the discriminating properties of the instrument, with
figures were 1551 (SD 551) and 978 (SD 497). The overall expected higher levels of self-efficacy and performance in
intra-class correlation coefficient was 095 (95% CI = 088- SMS. Discriminating properties at self-efficacy level were
098) for the self-efficacy items and 094 (95% CI = 085- not provided for masters educated nurses; yet they demon-
098) for the performance items. The intra-class correlation strated a markedly higher performance than non-master-
coefficients for the subscales are displayed in Figure 1. educated nurses. Master-educated nurses are supposed to
possess the reflective and critical thinking abilities needed in
Floor and ceiling effects more complex care settings (ter Maten-Speksnijder et al.
Table 4 presents the proportions of nurses scoring in the 2012). A more reflective attitude on professional perfor-
125% lower and upper bound, respectively, of (sub)scale mance is desirable, but can make persons more stringent in
values. Floor or ceiling effects were not found, apart from a judging their self-efficacy (Desmedt 2004, Koole et al.
Self-efficacy items
Nurses providing consultations (A) 60 1871 (381) 1675 (292) 370 7890 <0001
vs. Nurses on hospital units (B) 219
Nurses (A) 352 1722 (322) 1606 (383) 221 394 003
vs. Nursing students (B) 44
Nurses with a master degree (A) 59 1748 (368) 1707 (326) 094 366 035
vs. nurses without master degree (B) 309
Nurses perceiving SMS highly important* (A) 87 1775 (305) 1624 (309) 210 108 004
vs. nurses perceiving SMS of little to no importance (B) 23
Performance items
Nurses providing consultations (A) 60 1399 (436) 1147 (331) 417 7858 <0001
vs. Nurses on hospital units (B) 219
Nurses (A) 352 1202 (374) 939 (397) 423 391 <0001
vs. Nursing students (B) 41
Nurses with a master degree (A) 59 1300 (443) 1154 (370) 238 7416 002
vs. nurses without master degree (B) 306
Nurses perceiving SMS highly important* (A) 87 1260 (326) 1133 (267) 173 108 002
vs. nurses perceiving SMS of little to no importance (B) 23
*Score 9.
Score 6.
t, value independent sample t-test; d.f., degrees of freedom.
Table 4 Subscale and scale scores, including floor and ceiling effects (%).
Self-efficacy Performance
% % % %
Mean SD Min Max Mean SD Min Max
Subscale Assess* 296 063 040 1190 205 078 160 250
Subscale Advise 294 061 020 1270 205 071 100 150
Subscale Agree 269 074 110 620 166 086 720 130
Subscale Assist 281 067 000 1120 190 082 260 210
Subscale Arrange 251 079 020 760 149 085 1090 100
Subscale Overall Competencies 308 056 000 1600 253 073 000 450
Total scale 1709 331 000 1110 1175 384 000 100
2012). This might explain why masters educated nurses between both countries, rather than predefined group differ-
performed better, while being more prudent in the confi- ences.
dence of their own capacities. The small proportion of mas- The evidence to support the internal consistency of the
ter-educated nurses, whereby equal variance between instrument and its sub-scales was strong. The high Cron-
groups could not be assumed at performance level, may bachs alpha values, ranging from 075-096, indicate a
also explain these unexpected results. Nevertheless, some good to very good internal consistency or homogeneity
between-group differences could be the result of insufficient for the instrument and for the subscales. The results of
variation in professional status (nurses vs. students) the testretest procedure indicate that the stability of the
between the country samples and thereby reflect differences instrument was good, as the intra-class correlations
in conceptualization and implementation about SMS reached the recommended values 070. Hardly any floor
or ceiling effects were found, indicating the possibility to students in this study. To ensure international validity we
distinguish between individuals and to measure encourage initiatives to translate the SEPSS instrument into
changes after intervention. Attention is needed on the other languages and to validate it for use in the respective
estimation of self-efficacy for the Overall Competencies, countries.
reflecting the self-efficacy towards having a partnership
attitude, as an effect might be missed due to a possible
Limitations
ceiling effect.
The SEPSS is an instrument that captures nurses per- The study had some limitations. First, the low response rate
formance and self-efficacy in performing SMS. Given the in some subsamples and the lack of knowledge on the reasons
importance of self-efficacy as a precursor for behaviour for drop-out during the online completion of the question-
(Bandura 1991), we strongly recommend to assess the naire, might limit the generalizability of the findings. Never-
performance and self-efficacy items in an integrated way, theless, we were able to recruit a heterogeneous sample from
so as to make it feasible to work simultaneously on both different settings, representing nurses with and without expe-
areas where needed. The division in the six subscales rience in SMS and from two different countries, each having
enables to measure outcomes on subscale level and to a different history about self-management. This heterogene-
focus on a particular aspect of the SMS-process, while ity may have enhanced the representativeness of the sample.
the total score presents a more overall view of how SMS Second, the testretest procedure was performed in a small
is provided. Scores range from 0-4 for the subscales and group and the intensive procedure may have adversely
from 0-24 at total scale level. Higher scores on the affected attention during completion of the retest. Besides,
SEPSS instrument reflect a higher level of self-efficacy or the short time interval could have inflated the ICC values by
performance in SMS. the recall of the statements, although this seems not so obvi-
As the format of the SEPSS instrument requires nurses to ous for a comprehensive tool. Therefore, the results of the
rate both self-efficacy and performance on the same set of stability tests should be considered an initial trend. Further
items, a high correlation between both was not unimagin- stability testing in a larger sample is recommended. Third, by
able in view of the possibility of maintaining some coher- measuring at one point in time, we were not yet able to estab-
ence and consistency in responses. However, the response lish the instruments sensitivity to change in competence
patterns for self-efficacy and performance differed mark- development, which is one of the proposed applications. In
edly, as evidenced by the moderate correlation (r = 063, the future, we intend to use the SEPSS to measure the effect
P < 0001) found. of SMS training.
The instrument has several potential applications for
healthcare settings shifting towards SMS. First, the assess- Conclusion
ment of current SMS practice from a self-reported perspec-
tive, which may bring to light competencies that require In view of its good psychometric properties, the new SEPSS
training at an individual or department level. Second, this instrument is a promising instrument to measure nurses
assessment can help trainers in tailoring the content and self-efficacy and performance with regard to SMS. The self-
teaching strategies of training courses. Third, but this is a reported results could serve as an outcome measure of SMS
more reflective application, making nurses aware of possible practices in clinical and research settings, to identify educa-
discrepancies between their confidence and their perfor- tional needs and to evaluate personal growth and to assess
mance and the causes of these discrepancies. Fourth, train- the effectiveness of training or other interventions to
ing effectiveness and personal growth through training can improve SMS.
be evaluated, and the effectiveness of other interventions
aimed at improving SMS competencies. However, the Acknowledgements
instruments sensitivity to change has not yet been estab-
lished. Fifth, the total scale score could be useful to monitor We express our thanks to Claudia Gamel and Ko Hagoort
fidelity of SMS implementation. for the English translation of the SEPPS. We thank the
Considering that SMS is the responsibility of a multidisci- experts who were consulted in the development process of
plinary team whose members are expected to possess the the instrument. We also thank the Belgium and Dutch
same competencies (Wagner et al. 2001), it is recommended nurses and nursing students who participated in this study
to investigate the psychometric characteristics in groups of by completing the questionnaire. We are grateful to Ko for
other healthcare professionals than the nurses and nursing linguistical and editorial support.
Kim S.C., Boren D. & Solem S.L. (2001) The Kim Alliance Scale: interviewing. Journal of Substance Abuse Treatment 28(1), 19
development and preliminary testing. Clinical Nursing Research 26.
10(3), 314. Nolte E. & McKee M. (2008) Caring For People With Chronic
Koole S., Dornan T., Aper L., Scherpbier A., Valcke M., Cohen- Conditions: A Health System Perspective: A Health System
Schotanus J. & Derese A. (2012) Does reflection have an effect Perspective. McGraw-Hill International, Berkshire.
upon case-solving abilities of undergraduate medical students? Polit D.F. & Beck C.T. (2008) Nursing Research: Generating and
BMC Medical Education 12(1), 75. Assessing Evidence for Nursing Practice. Lippincott Williams &
Kosmala-Anderson J., Wallace L.M. & Turner A. (2010a) Does the Wilkins, Philadelphia.
professional and working context of United Kingdom clinicians Pols R.G. (2009) Chronic condition self-management support:
predict if they use practices to support patients with long term proposed competencies for medical students. Chronic Illness 5
conditions to self manage. Archives of Medical Science 6, 815 (1), 714. doi:10.1177/1742395308098888.
821. Sandman L., Granger B.B., Ekman I. & Munthe C. (2012)
Kosmala-Anderson J.P., Wallace L.M. & Turner A. (2010b) Adherence, shared decision-making and patient autonomy.
Confidence matters: a Self-Determination Theory study of factors Medicine, Health Care and Philosophy 15(2), 115127.
determining engagement in self-management support practices of doi:10.1007/s11019-011-9336-x.
UK clinicians. Psychology Health & Medicine 15(4), 478491. Scholl I., Kriston L., Dirmaier J., Buchholz A. & Harter M. (2012)
doi:10.1080/13548506.2010.487104. Development and psychometric properties of the Shared Decision
Kosmala-Anderson J., Wallace L.M., Turner A. & Barwell F. Making Questionnairephysician version (SDM-Q-Doc). Patient
(2011) Development and psychometric properties of a self report Education and Counseling 88(2), 284290. doi:10.1016/
measure to assess clinicians practices in self management j.pec.2012.03.005.
support for patients with long term conditions. Patient Schulman-Green D., Jaser S., Martin F., Alonzo A., Grey M.,
Education and Counseling 85(3), 475480. doi:10.1016/ McCorkle R. & Whittemore R. (2012) Processes of self-
j.pec.2010.10.007. management in chronic illness. Journal of Nursing
Krichbaum K., Aarestad V. & Buethe M. (2003) Exploring the Scholarship 44(2), 136144. doi:10.1111/j.1547-5069.2012.
connection between self-efficacy and effective diabetes self- 01444.x.
management. The Diabetes Educator 29(4), 653662. Stacey D., Taljaard M., Drake E.R. & OConnor A.M. (2008)
Kriston L., Scholl I., H olzel L., Simon D., Loh A. & Harter M. Audit and feedback using the brief Decision Support Analysis
(2010) The 9-item Shared Decision Making Questionnaire Tool (DSAT-10) to evaluate nursestandardized patient
(SDM-Q-9). Development and psychometric properties in a encounters. Patient Education and Counseling 73(3), 519525.
primary care sample. Patient Education and Counseling 80(1), doi:10.1016/j.pec.2008.07.016.
9499. doi:10.1016/j.pec.2009.09.034. Terwee C.B., Bot S.D., de Boer M.R., van der Windt D.A., Knol
Lawn S., Battersby M., Lindner H., Mathews R., Morris S., Wells D.L., Dekker J. & de Vet H.C. (2007) Quality criteria were
L. & Reed R. (2009) What skills do primary health care proposed for measurement properties of health status
professionals need to provide effective self-management support? questionnaires. Journal of Clinical Epidemiology 60(1), 3442.
Seeking consumer perspectives. Australian Journal of Primary Thorne S.E., Ternulf Nyhlin K. & Paterson B.L. (2000) Attitudes
Health 15(1), 3744. doi:10.1071/PY08053. toward patient expertise in chronic illness. International Journal
Lawn S., Delany T., Sweet L., Battersby M. & Skinner T.C. (2013) of Nursing Studies 37(4), 303311. doi:10.1016/S0020-7489(00)
Control in chronic condition self-care management: how it 00007-9.
occurs in the health worker-client relationship and implications Udlis K.A. (2011) Self-management in chronic illness: concept and
for client empowerment. Journal of Advanced Nursing 70(2), dimensional analysis. Journal of Nursing and Healthcare of
383394. doi:10.1111/jan.12203. Chronic Illness 3(2), 130139. doi:10.1111/j.1752-9824.2011.
Leisen B. & Hyman M.R. (2001) An improved scale for assessing 01085.x.
patients trust in their physician. Health Marketing Quarterly 19 Vallis M. (2013) Behaviour change counselling how do i know if
(1), 2342. doi:10.1300/J026v19n01_03. i am doing it well? The development of the Behaviour Change
Lorig K.R. & Holman H. (2003) Self-management education: Counselling Scale (BCCS). Canadian Journal of Diabetes 37(1),
history, definition, outcomes and mechanisms. Annals of 1826.
Behavioral Medicine 26(1), 17. doi:10.1207/S15324796 Visse M.A., Teunissen T., Peters A., Widdershoven G.A.M. &
ABM2601_01. Abma T.A. (2010) Dialogue for air, air for dialogue: towards
ter Maten-Speksnijder A.J., Grypdonck M.H., Pool A. & Streumer shared responsibilities in COPD practice. Health Care Analysis
J.N. (2012) Learning opportunities in case studies for becoming 18(4), 358373. doi:10.1007/s10728-009-0139-7.
a reflective nurse practitioner. Journal of Nursing Education 51 Wagner E.H., Austin B.T., Davis C., Hindmarsh M., Schaefer J. &
(10), 563569. doi:10.3928/01484834-20120820-05. Bonomi A. (2001) Improving chronic illness care: Translating
McDonald R., Rogers A. & Macdonald W. (2008) Dependence evidence into action. Health Affairs 20(6), 6478. doi:10.1377/
and identity: nurses and chronic conditions in a primary care hlthaff.20.6.64.
setting. Journal of Health Organization and Management 22(3), WHO (2014) Global Status Report on Noncommunicable
294308. doi:10.1108/14777260810883558. Diseases. Attaining the Nine Global Noncommunicable Diseases
Moyers T.B., Martin T., Manuel J.K., Hendrickson S.M. & Miller Targets; a Shared Responsibility. World Health Organization,
W.R. (2005) Assessing competence in the use of motivational Geneva.
Wilson P.M., Kendall S. & Brooks F. (2006) Nurses responses to study. Patient Education and Counseling 90(1), 2937.
expert patients: the rhetoric and reality of self-management in doi:10.1016/j.pec.2012.09.003.
long-term conditions: a grounded theory study. International Zwar N., Harris M., Griffiths R., Roland M., Dennis S., Powell
Journal of Nursing Studies 43(7), 803818. doi:10.1016/ Davies G. & Hasan I. (2006) A Systematic Review of Chronic
j.ijnurstu.2005.10.011. Disease Management. Research Centre for Primary Health Care
Yank V., Laurent D., Plant K. & Lorig K. (2013) Web-based self- and Equity, School of Public Health and Community Medicine,
management support training for health professionals: a pilot UNSW, New South Wales.
The Journal of Advanced Nursing (JAN) is an international, peer-reviewed, scientific journal. JAN contributes to the advancement of
evidence-based nursing, midwifery and health care by disseminating high quality research and scholarship of contemporary relevance
and with potential to advance knowledge for practice, education, management or policy. JAN publishes research reviews, original
research reports and methodological and theoretical papers.
For further information, please visit JAN on the Wiley Online Library website: www.wileyonlinelibrary.com/journal/jan
High-impact forum: the worlds most cited nursing journal, with an Impact Factor of 1527 ranked 14/101 in the 2012 ISI Jour-
nal Citation Reports (Nursing (Social Science)).
Most read nursing journal in the world: over 3 million articles downloaded online per year and accessible in over 10,000 libraries
worldwide (including over 3,500 in developing countries with free or low cost access).
Fast and easy online submission: online submission at http://mc.manuscriptcentral.com/jan.
Positive publishing experience: rapid double-blind peer review with constructive feedback.
Rapid online publication in five weeks: average time from final manuscript arriving in production to online publication.
Online Open: the option to pay to make your article freely and openly accessible to non-subscribers upon publication on Wiley
Online Library, as well as the option to deposit the article in your own or your funding agencys preferred archive (e.g. PubMed).