Applied Nursing Research 27 (2014) 33–40

Contents lists available at ScienceDirect

Applied Nursing Research journal homepage: www.elsevier.com/locate/apnr

Quality of Care and Its Impact on Quality of Life for Care-Dependent Persons With Dementia in Shared-Housing Arrangements: Results of the Berlin WGQual-Study Karin Wolf-Ostermann, PhD, Dipl.-Statistikerin ⁎, Andreas Worch, Dipl. Pflegewirta, Saskia Meyer, MSc, Johannes Gräske, Dipl. Pflegewirt Alice Salomon Hochschule Berlin - University of Applied Sciences; Alice Salomon Platz 5, D- 12627 Berlin, Germany

a r t i c l e

i n f o

Article history: Received 23 January 2013 Revised 29 October 2013 Accepted 1 November 2013 Keywords: Dementia Quality of care Quality of life Shared-housing arrangements Small scale living

a b s t r a c t Aim: The aim of study was to evaluate setting-specific quality indicators (QIs) for shared-housing arrangements (SHA) regarding effects of a guided quality development process on resident's quality of life (QoL). Background: SHA are a specific German kind of small-scale living facilities for care-dependent persons with dementia. SHA are disconnected from residential facilities and served by community care services. Method: In a longitudinal cluster-randomised design 104 residents of 34 SHA were surveyed for 1 year; the intervention group took part in a quality development process. QIs, physical and psychological health outcomes including QoL were surveyed. Results: During the 1 year follow-up, analyses show a positive trend regarding QI-outcomes in the intervention group, but no statistically significant differences could be proved regarding QoL. Conclusions: Setting-sensitive QIs are absolutely essential to measure quality of care in multi-professional settings like SHA. Further research as well as longer study intervals is essential. © 2014 Elsevier Inc. All rights reserved.

Introduction The number of people with dementia (PwD) is rapidly increasing all over the world (Wimo & Prince, 2010). Results of Marquardt and Schmieg (2009) show that the design of the living environment has an impact on QoL and on behavioural problems of PwD. In general, there is no consensus on the supremacy of one special care setting in terms of beneficial aspects (e. g. QoL, challenging behaviour, …) for PwD (Kane, Lum, Cutler, Degenholtz, & Yu, 2007; te Boekhorst, Depla, de Lange, Pot, & Eefsting, 2009, Verbeek et al. (2010), Wolf-Ostermann, Worch, Fischer, Wulff, & Gräske (2012)). German SHA were developed in line with the international trend of of small-scale, homelike living facilities being an alternative care arrangement to traditional care facilities (Wolf-Ostermann, et al., 2012). Typically, six to eight residents share one large apartment in a mostly urban setting (Wolf-Ostermann, et al., 2012, Gräske et al., 2013). SHA are completely disconnected from residential care, being served by community care service providers (Wolf-Ostermann & Fischer, 2010). SHA are characterised by different groups of relevant stakeholders: residents, relatives, volunteers, care provider, physicians, therapists and landlords. Main objectives of SHA are the maintenance of self-determination despite of high needs in care and ⁎ Corresponding author at: Alice Salomon Hochschule Berlin - University of Applied Sciences; Alice Salomon Platz 5, D- 12627 Berlin, Germany. Tel.: +49 30 99245 507; fax: +49 30 99245 245. E-mail address: [email protected] (K. Wolf-Ostermann). 0897-1897/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.apnr.2013.11.002

providing a life as “normal” as possible in family-like structures, the involvement of family members in daily living, the connection to neighborhood, ensured access to care and well-being, and obtaining self-esteem and autonomy (Wolf-Ostermann, et al., 2012, Gräske et al., 2013). Within the last decade, the number of SHA in Germany for care-dependent PwD showed a tenfold increase from 143 SHA in 2003 to more than 1,500 SHA with ca. 11,000 residents in 2012 (WolfOstermann, Worch, Meyer, & Gräske, 2013). In contrast to residential care no special legal guidelines for SHA in Germany concerning quality of care and support exist besides general legal quality requirements for care provider. SHA are judged as private domesticities although PwD are considered to be a vulnerable group of people who have to be especially protected. Since different settings have different requirements in terms of providing high quality care for PwD, quality development needs to be tailored to given structural (e. g. number and qualification of staff) and procedural (e. g. nursing interventions) conditions. For long-term care in nursing homes, Nakrem, Vinsnes, Harkless, Paulsen, & Seim (2009) give an overview of nursing sensitive quality indicators (QIs). In the case of dementia care QIs “mainly focus on diagnostics, screening, and pharmacological interventions” (Vasse et al., 2012). For small-scale living facilities like the German SHA, particular criteria for assessing and appraising the quality of care beyond the well-established indicators such as falls, pressure ulcer, nutrition etc., are widely lacking. Based on a systematic literature review, a set of 39 QIs was developed, taking into account the multi-professional network of care providers in SHA and focusing on different aspects of structures,

34

K. Wolf-Ostermann et al. / Applied Nursing Research 27 (2014) 33–40

procedures and outcomes (see table 1) (Worch, Gräske, Dierich, & Wolf-Ostermann, 2011). The concept behind is to establish QIs which are easy to measure but are nevertheless evidence-based and not solely on expert opinion. A set of QIs meeting these requirements can be used routinely by care providers to enhance their quality of services as well as by PwD and their informal caregivers to compare SHA by means of quality aspects. The aim of the present publication is to evaluate a set of previously developed QIs tailored to SHA in order to ascertain if a guided quality development process by means of these QIs leads to better health outcomes and better quality of life (QoL) for the residents. Methods Design A cluster-randomised longitudinal design with measurements before intervention (t1) and twelve months (t2) later was used to evaluate a set of QIs for measuring quality of care and support in SHA and outcomes of residents in SHA in Berlin, Germany. In a sequential two-step recruiting process first care services in SHA were recruited and afterwards residents of these SHA. The intervention group consisted of SHA which were undergoing an externally guided quality development process; the control group proceeded without this. Ethical approval The study was approved by the Ethics Committee of the German Society of Nursing Science (declaration of consent 28th of May 2010). Setting In 2010, existing community care provider in Berlin/Germany providing services in SHA were informed about the purpose of the study and invited to participate; 11 out of 132 care service providers and 37 out of 400 SHA expressed interest in participating in the study. Main reasons for not participating in the study were shortage of resources and lack of interest. Sample All 261 potential study participants were informed about the aims of our study in April 2010. The total sample at baseline consisted of 34 SHA and 104 participants (39.8% of all residents, see Fig. 1). All participants or their legal representatives signed the informed consent to be included into the study. During follow-up, 36 (34.6%) mostly older and more cognitively impaired persons dropped out of the study. A total of 68 persons participated in all measurements during 1-year follow-up period. A power calculation (G-Power v.3.0.10) for a repeated measurements ANCOVA for our main outcome parameter QoL (QUALIDEM, 10 subscales) with two groups, a Bonferroni-corrected α = 0.005 and power (1-ß) = 0.95 yielded a total sample size n = 44. We assumed a medium effect size (d = 0.36), which corresponds to a difference in five points on the QUALIDEM scales (0–100). Outcome measures A set of 39 evidence-based QIs was previously developed especially for SHA (Worch, et al., 2011) and was evaluated. Residents' QoL as well as ADL-functioning, cognitive impairment, need-driven behaviour and demographical factors were surveyed. The main outcome measure was QoL (QUALIDEM). Study nurses (registered nurses (RN)) were trained in conducting research interviews and assessed the QIs as well as residents' outcomes. Information was obtained using general documentation, nursing records and face-to-

face interviews blinded to group assignment. Proxy-ratings were performed by asking the staff. Quality indicators Quality requirements concerning all service providers in SHA are not wide spread. These are necessary for a sufficient care for the residents, which are predominantly with dementia. To identify care aspects of these people, which improve cognitive and functional capacity, quality of life or neuropsychiatric symptoms, we conducted a systematic literature research in databases CareLit ®, CINAHL ®, GeroLit ® and PubMed ® (Worch et al. (2011)) including 567 publications. Based on this, we developed a set of QIs for measurement and improvement of quality. The QIs include relevant aspects of structures, processes, outcomes (see table 1) and involve all service providers in SHA (nursing staff, medical practitioners, therapists, family caregivers, volunteers and landlords). Every QI - except the ratio of residents per staff persons in SHA - is expressed as the percentage of the available total number. The QIs partly refer to the setting SHA as a whole and partly in detail to all residents per SHA. Some QIs, focusing on service provision of health care providers, were only evaluated in the intervention group (see table 1) in order not to initiate quality development processes based on the QIs in the control group. A summative score for each resident was calculated counting the percentage of QIs being fulfilled. This QI-total-score is standardized from 0–100; all QIs were equally weighted. In order to measure the overall quality we recoded all available QI into dichotomous variables, with a value of one indicating a quality above average and a value of zero indicating a quality below average. The relative QI total score (0–100) was calculated by summing up the recoded variables and dividing the sum by the number of contributing QIs. Cognitive impairment The internationally validated performance-based instrument mini mental state examination (MMSE) was used to screen cognitive status (Folstein, Folstein, & McHugh, 1975). The MMSE contains 30 questions with a total range from 0–30 points, greater values indicating better cognitive abilities. To assess the severity of dementia, the Global Deterioration Scale (Reisberg, Ferris, de Leon, & Crook, 1982) was applied to grading residents into stages 1 (objectively normal person) to 7 (severe dementia). Higher values indicate an increasing stage of dementia. The instrument is both valid and reliable. Neuropsychiatric symptoms The presence of physical aggressive behaviour, physical nonaggressive behaviour and verbally agitated behaviour were measured using the Cohen–Mansfield agitation inventory (CMAI, Cohen-Mansfield, Marx, & Rosenthal, 1989), with proved validity and reliability. Activities of daily living Activities of daily living were measured by the Extended Barthel Index (EBI, Prosiegel et al., 1996), which shows sufficient validity and reliability. Sixteen variables describing activities of daily living and cognitive abilities are summed up to a total score ranging from 0 (completely dependent) to 64 (independent). Quality of life To evaluate QoL, the QUALIDEM was used (Ettema, Dröes, Lange, Mellenbergh, & Ribbe, 2007), which is especially applicable in SHA. Currently, the QUALIDEM is the best available instrument to assess (in terms of acceptance, reliability and validity) residents' QoL in SHA (Gräske, et al., 2013). The QUALIDEM is a proxy-rating instrument,

Table 1 Quality indicators (QIs) for SHA in the intervention and control group for residents with 1-year follow-up (n = 68): ▲ Improvement over time, ▼ deterioration over time, = unchanged values. Quality Indicator

Based on

t1

t1

t2

Change over time/p

1.3 29.7% 5.0% 9.1% 70.0% 91.8% 66.7%

0.6 15.4% 5.2% 7.6% 88.9% 91.8% 66.7%

▲/0.006⁎ ▼/0.160 ▲/0.347 ▼/0.605 ▲/0.582 = =

90.9% 90.9%

90.9% 90.9%

= =

SHA SHA SHA SHA Residents SHA SHA Residents Residents SHA SHA Residents Residents Residents Residents Residents

100.0% 100.0% 100.0% 100.0% 54.8% 100.0% 33.3% 100.0% 91.8% 20.0% 100.0% 100.0% 12.9% 45.2% 6.5% 0.0%

100.0% 100.0% 100.0% 100.0% 35.5% 100.0% 0.0% 100.0% 100.0% 77.8% 100.0% 93.5% 19.4% 41.9% 22.6% 3.2%

= = = = ▼/0.013⁎ = ▼/0.076 = ▲/0.145 ▲/0.023⁎

100.0% 100.0% 96.4% 100.0% 51.4% 100.0% 9.5% 100.0% 100.0% 89.2% 2.7% 48.6% 0.0% 2.7%

100.0% 100.0% 100.0% 100.0% 43.2% 100.0% 19.0% 100.0% 100.0% 94.6% 10.8% 59.5% 10.8% 2.7%

= = ▲/0.290 = ▼/0.054 = ▲/0.003⁎ = = ▲/0.674 ▲/0.182 ▲/0.297 ▲/0.050⁎

Residents Residents Residents Residents Residents Residents Residents Residents Residents Residents Residents Residents Residents Residents Residents Residents

23.3% 3.2% 9.7% 38.7% 16.1% 16.1% 12.9% 74.2% 3.2% 45.2% 41.9% 35.5% 9.7% 96.8% 80.6% 93.5%

16.7% 6.5% 3.2% 35.5% 12.9% 22.6% 29.0% 64.5% 6.5% 41.9% 29.0% 19.4% 19.4% 100.0% 90.3% 96.8%

▲/0.768 ▼/1.000 ▲/0.612 ▲/1.000 ▲/1.000 ▼/0.749 ▼/0.211 ▼/0.065 ▲/0.319 ▼/0.990 ▼/0.076 ▼/0.080 ▲/0.315 ▲/1.000 ▲/0.010⁎

20.6% 8.1% 10.8% 43.2% 16.2% 24.3% 10.8% 59.5% 13.5% 43.2% 59.5% 29.7% 2.7% 94.6% 91.9% 89.2%

26.5% 2.7% 16.2% 43.2% 10.8% 24.3% 13.5% 70.3% 5.4% 40.5% 29.7% 16.2% 13.5% 97.3% 83.8% 78.4%

▼/0.945 ▲/0.615 ▼/0.736 = ▲/0.736 = ▼/1.000 ▲/0.590 ▼/0.266 ▼/0.811 ▼/0.080 ▼/0.523 ▲/0.199 ▲/1.000 ▼/0.046⁎ ▼/0.345

Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of

residents per staff persons in SHA# registered nurses per total nursing staff in SHA★ gerontologically advanced registered nurses in SHA★ nursing staff with special psychogeriatric qualification in SHA★ SHA providing special staff education concerning “dementia”★ SHA with outdoor facility★ SHA with unlocked outdoor facility regarding all SHA with outdoor facilities★

SHA SHA SHA SHA SHA SHA SHA

Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of

SHA with relational care★ SHA assessing biographical data of residents★ SHA assessing and updating biographical data of residents at regular intervals★ SHA using biographical data of residents for care and support★ residents with relatives which are participating in care and support★ SHA providing information and advisory services for family caregivers at regular intervals★ SHA with volunteers participating in care and support★ residents getting reminiscence therapy★ residents receiving care based on techniques of “basal stimulation“★ SHA using the concept of validation★ SHA providing meaningful group activities★ residents participating in meaningful group activities★ residents getting external speech therapy within the last 12 months by medical prescription★ residents getting external physical therapy within the last 12 months by medical prescription ★ residents getting external occupational therapy within the last 12 months by medical prescription★ residents getting external music therapy within the last 12 months by medical prescription★

Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of Ratio of

residents having underweight (BMI b 24.0 kg/m2)# residents having pressure ulcers# residents being fallen within the last 4 weeks # residents with verbal agitation# residents with aberrant motor behaviour (clinical relevance) residents being verbally or physically aggressive# apathetic residents (clinical relevance) # residents with regular social contacts★ residents being able to wash themselves★ residents being able to use the toilet without help★ residents being able to eat without help ★ residents with sufficient mobility ★ residents using acute care★ residents using routine primary care★ residents using routine specialist care★ residents without physical restraints★

#

t2

Control group study results at

= ▼/0.492 ▲/0.431 ▼/0.184 ▲/0.158 ▲/0.001

▲/1.000

=

K. Wolf-Ostermann et al. / Applied Nursing Research 27 (2014) 33–40

Structures QI S1 QI S2 QI S3a QI S3b QI S4 QI S5a QI S5b Procedures QI P1 QI P2a QI P2b QI P2c QI P3a QI P3b QI P4 QI P5 QI P6 QI P7 QI P8a QI P8b QI P9a QI P9b QI P9c QI P9d Outcomes QI E1 QI E2 QI E3 QI E4a QI E4b QI E4c QI E4d QI E5 QI E6a QI E6b QI E6c QI E6d QI E7a QI E7b QI E7c QI E8

Change over time/p

Intervention group study results at

⁎ Significant at p b .05. ★ Higher ratios indicate a better quality of care. # Lower ratios indicate better quality of care.

35

36

K. Wolf-Ostermann et al. / Applied Nursing Research 27 (2014) 33–40

consisting of 37 single items measured on a 4-point scale which can be summarized to nine subscales and one total scale (range 0–100): “care relationship” (7), “positive affect” (6), “negative affect” (3), “restless tense behaviour” (3), “positive self-image” (3), “social relations” (6), “social isolation” (3), “feeling at home” (4), “having something to do” (2). Increasing values in all scales correspond with increasing QoL. According to Bouman et al. (2011) people with severe dementia (Global Deterioration Scale = 7 (Reisberg, et al., 1982)) are excluded from the analysis of three domains (“positive self-image”, “feeling at home”, “having something to do”). External-guided quality development Quality of care and support was evaluated by the standardized set of QIs at t1 and t2. With the help of an external trained (European Foundation for Quality Management, European Quality Organisation) quality manager, in each SHA in the intervention group the particular results were discussed. Based on the measured QIs the external quality manager guided the staff of SHA to develop a tailored quality concept for each SHA and to operate this process more and more on their own. Depending on the particular results, agreed objectives and activities, the guided quality development process in each SHA was tailored individually to given conditions of structural and procedural features. In the beginning of the quality development process monthly meetings in each SHA in the intervention group were held. Relevant groups (staff, relatives, volunteers) attended these meetings, which were supervised by the quality manager. In the course of the study the frequency of meetings was extended up to 3 months. This approach was used to ensure a continuous quality development even after the study period. Additionally staff members in the intervention group were taught during follow-up how to assess the QIs by themselves in order to give a long-term perspective. Statistical analysis Basic characteristics of residents were described using descriptive statistics. Correlations among metric/ordinal variables were examined

by Pearson's and Spearman's correlations. Chi-square tests, Fisher's exact test and t-tests were also used to analyse the data. Fixed-effects models of analysis of co-variance (ANCOVA) were used to analyse longitudinal effects of the major outcome parameters (QUALIDEM subscales). These models were adjusted for confounding factors such as sex, age, length of stay, stage of dementia (GDS), occurrence of at least one challenging behaviour (CMAI) and values of the outcome parameter at baseline. Interactions between confounding variables were not modelled because of the small number of study participants. Statistical model assumptions of normal distribution and multicollinearity for variables were examined before conducting further analyses. Significance (nominal p value) was set at p b .05; and p b .10 was defined to be significant by trend. For the longitudinal analyses, a Bonferroni-correction was applied (α = 0.005) to the overall level of significance, and only measurements of n = 68 participants with complete data for the 1-year follow-up period were used. All statistical analyses were carried out using SPSS® (v20.0). Results Quality indicators The study nurses easily obtained the developed QIs from staff members in SHA. They instructed the staff carefully how to assess the relevant data. No misunderstandings occurred during the study period. The QIs were evaluated by the staff as being feasible. Ten SHA with a total of 49 residents at baseline (intervention group) participated in the quality development process (see table 1, Fig. 1). Concerning structures, a ratio of about 1.3 residents per one SHA staff member was observed at baseline, with about 30% of registered nurses per total nursing staff, but only 5% being certified in psychogeriatric nursing. The ratio of staffing in total increased significantly to 0.6 residents per staff persons during 1 year of follow-up. The percentage of staff with a special education regarding “dementia” increased slightly from 70.0% to 88.9%. Concerning procedures, all SHAs obtained biographical data of their residents. All SHAs provided meaningful group activities, education and

Fig. 1. Flow charts of participants.

K. Wolf-Ostermann et al. / Applied Nursing Research 27 (2014) 33–40

advisory services for family caregivers at regular intervals. The ratio of residents with relatives actively participating in the care and support of the resident decreased significantly from 54.8% (t1) to 35.5% (t2) as well as the ratio of volunteers did (t1: 33.0%, t2: 0%). There was an upward trend on the percentage of residents receiving special external therapies (speech, occupational, and music) except for external physical therapy which decreased slightly (t1: 45.2%, t2: 41.9%). Also the ratio of SHA using the concept of validation increased significantly (t1: 20.0%, t2: 77.8%). Outcomes: Outcomes which are closely related to ongoing dementia and therefore decline of functional abilities and those which can be influenced by care and support have to be distinguished. The percentage of residents being able to eat without help or ambulate with sufficient mobility decreased significantly during the 1 year follow-up. The percentage of residents with regular social contact also declined significantly (t1: 74.2%, t2: 64.5%). In addition, the percentage of residents exhibiting clinically relevant apathy increased as well as resident's utilization of acute care services. In contrast, a positive development in using routine primary or specialist care and decreased use of physical restraints as residents exhibited challenging behaviours was noted. Relevant care-related outcomes such as the percentage of residents, who were underweight or have suffered a fall in the last 4 weeks prior to data collection, also decreased. The QI-total-score for the intervention group reached 58.7 (s = 6.3) at t1 and 57.4 (s = 4.9) at t2, remaining almost unchanged. In the control group, values of 60.3 (s = 7.4) at t1 and 56.7 (s = 8.8) at t2 were achieved, but in this group a few QIs were not measured. The

37

decrease of the QI-total-score in the control group was statistically significant (paired t-test, p = .008). Participants A total of 104 residents (76 female, 28 male, average age 79.0 years) were recruited into the study (Fig. 1), 68 of them could be followed for the 1-year study period with complete measurements. The participants lived in SHA on average for nearly 3 years (33 months). Eighty-seven percent had a diagnosis of dementia with moderate level of cognitive impairment (mean MMSE = 11.5). The study population on average exhibited limitations in their functional abilities. The mean EBI for all participants (n = 104) at baseline was 32.9 (intervention group: 33.4. control group: 32.2). No statistically significant differences between intervention (n = 49; 47.1%) and control group (n = 55; 52.9%) could be established at baseline except length of stay and two dimensions of QoL (“negative affect” and “feeling at home”). Participants of the intervention group showed an average length of stay that is 1 year shorter. Regarding QoL, residents in the intervention group have on average a lower QoL for “negative affect” (−11 points) and “feeling at home” (− 16 points) (Table 2). Cognitive impairment The mean MMSE score of all participants (n = 104) at baseline was 11.5 (intervention group 11.3. control group: 11.9). More than

Table 2 Characteristics of follow-up residents (n = 68) at baseline (t1) and after 1-year-follow-up (t2). Population with complete follow-up: n = 68 Baseline t1

Sex in % (n) Female Male Age in years M(SD) Length of stay in years M(SD) Care level⁎ at t1 in % (n) None I II III MMSE M (SD) GDS in % (n) ≤5 6 7 CMAI in % (n) Physical non-agressive Verbal agitation Physical aggressive QUALIDEM M (SD)⁎⁎ Total score Care relationship Positive affect Negative affect Restless tense behaviour Positive self-image⁎⁎⁎ Social relations Social isolation Feeling at home⁎⁎⁎ Having something to do⁎⁎⁎

p

Intervention group(n = 31)

Control group(n = 37)

74.2 (23) 25.8 (8) 77.5 (9.1) 3.3 (2.2)

78.4 21.6 77.8 2.4

(29) (8) (10.9) (1.8)

0.0 (0) 19.4 (6) 51.6 (16) 19.4 (6) 11.3 (9.9)

2.7 18.9 62.2 16.2 12.7

(1) (7) (23) (6) (9.8)

12.9 (4) 51.6 (16) 35.5 (11)

10.8 (4) 54.1 (20) 35.1 (13)

38.7 (12) 38.7 (12) 16.1 (5)

62.2 (23) 43.2 (16) 24.3 (9)

67.3 67.3 72.5 71.9 57.3 61.0 73.1 59.9 82.1 52.5

66.8 68.3 76.4 59.3 56.8 74.1 71.9 62.5 68.2 53.5

One-year follow-up t2

p

Intervention group(n = 31)

Control group(n = 37)

6.5 32.3 48.4 12.9 12.0

0.0 (0) 21.6 (8) 59.5 (22) 18.9 (7) 10.6 (9.2)

n.s.2

(17.2) (22.3) (22.3) (27.5) (32.1) (26.9) (18.1) (27.5) (14.1) (28.8)

(14.3) (23.1) (21.8) (30.1) (28.3) (23.3) (19.9) (22.1) (24.9) (30.3)

n.s.2 .0522 n.s.3

n.s.2 n.s.3

n.s.3 (2) (10) (15) (4) (10.0)

n.s.2 n.s.3

25.8 (8) 45.2 (14) 29.0 (9)

13.5 (5) 54.1 (20) 32.4 (12)

.0461 n.s.1 n.s.1

38.7 (12) 35.5 (11) 22.6 (7)

43.2 (16) 43.2 (16) 24.3 (9)

n.s1 n.s.1 n.s.1

n.s.2 n.s.2 n.s.2 .0792 n.s.2 .0922 n.s.2 n.s.2 .0332 n.s.2

68.7 69.9 73.9 71.3 59.1 61.6 71.7 64.9 90.2 48.5

70.4 67.3 83.9 70.9 58.9 64.0 69.2 66.4 80.3 57.3

n.s.2 n.s.2 .0312 n.s.2 n.s.2 n.s.2 n.s.2 n.s.2 .0372 n.s.2

(14.4) (19.2) (22.2) (20.9) (31.6) (31.0) (20.2) (24.2) (11.1) (36.7)

(12.8) (19.0) (15.1) (23.7) (30.4) (19.3) (22.0) (21.8) (18.8) (29.7)

n. s. = Not significant, M = arithmetic mean, SD = standard deviation. 1 Fisher's exact test. 2 t-Test. 3 Chi-square test. ⁎ Care level determined by the German long-term care insurance, level of care I, II, and III requires at least 90 minutes, 3 hours, and 5 hours of care per day, respectively. ⁎⁎ Range for all (sub)scales: 0–100. ⁎⁎⁎ Not applied for person with severe dementia (GDS = 7).

38

K. Wolf-Ostermann et al. / Applied Nursing Research 27 (2014) 33–40

half of the study population suffered from severe cognitive impairment (MMSE b 10). For study participants who completed the 1 year follow-up (n = 68, see table 2), analyses showed a significant decrease of the mean MMSE-score in the control group but an almost unchanged mean MMSE-score in the intervention group (all paired ttests; intervention group − 0.4. p = .834. control group: − 4.0. p = .006). Neuropsychiatric symptoms The occurrence of challenging behaviours did not differ significantly between intervention and control group at baseline and followup. The occurrence of challenging behaviours other than physical nonaggressive remained constant in the control group during follow-up while some (not significant) changes in the intervention group were observed (c.f. table 2). Quality of life At baseline, QoL of all participants (n = 104) reached moderate to high values (mean total score 67.4; intervention group: 66.6; control group: 68.4; t-test; p = .548). Most positive evaluations were reached for the dimensions “feeling at home” (74.7), “positive affect” (73.8) and “social relations” (71.4). QoL was rated less positively in the dimensions “restless tense behaviour” (58.7) and “having something to do” (54.0). No statistically significant differences were observed between intervention and control group in all dimensions of the QUALIDEM (all t-test; p N .05), except for “feeling at home” (ttest; p = .033) being higher in the intervention group. A slight but not significant increase of 2.6 points for the mean QUALIDEM total score during follow-up was observed (n = 68, see table 2). The highest increase was found for “feeling at home” (9.0 points; paired t-test; p = .020) and “negative affect” (6.1 points; paired t-test; p = .079). Including differences in sex, age, length of stay, stage of dementia (GDS), challenging behaviour and QI-totalscore as well as baseline values of the respective QUALIDEM scale into the ANCOVA no significant differences between intervention and control group over time could be shown (see table 3). The QUALIDEM total score as well as most of its subdomains at t2 were solely dependent on the QUALIDEM total score or its subdomains observed at baseline. Persons with high score values at t1 also had high score values at t2 (Table 3).

Discussion The main purpose of the study was to evaluate QIs which are tailored to the setting of SHA. Based on the high standard of a clusterrandomised study with a 1-year follow-up, the development of residents' QoL was examined to shed light on the question whether a guided quality development in SHAs tailored to given structural and procedural features shows substantial results in the defined outcome parameters even when we have to take into account that the progression of dementia is working against all improvements. Quality indicators The results reported by the staff of SHA in the intervention group show that the proposed QIs are in general applicable in SHA and can easily be obtained by nursing staff without the help of an external quality manager. One of the great advantages of using the proposed QIs in a continuing quality development process is that this quality development can be tailored to site-specific needs and without raising the cost of care. The ratio of residents to staff at baseline is comparable to results of a previous study by Wolf-Ostermann, Worch, Wulff, & Gräske (2011) in 105 SHAs. Regarding the follow-up, the ratio improved. But we cannot conclude that there was more staff employed in the SHA since it might also be possible that the number of residents decreased while the number of staff persons remained unchanged. The ratios of residents with relatives actively participating in care and support are lower than the results reported by Gräske et al. (2011). Wulff, Gräske, Fischer, & Wolf-Ostermann (2011) suggested that the utilization of routine primary and specialist care in SHAs should be improved. In the present study, a high percentage of utilization was noted at baseline and even increased during follow-up. The use of physical restraints in the present study decreased in the intervention group and is much lower than results reported by Gulpers et al. (2011) for only belt restraints at the beginning of the EXBELT-program in Dutch nursing homes (17%–19%). Results in the control group were approximately equal to those of Gulpers et al. (2011). The results emphasize the importance of quality developments concerning the use of physical restraints. Relevant care-related outcomes like the percentage of residents who are underweight or have suffered a fall decreased in the intervention group during follow-up. Meyer, Gräske, Worch, & WolfOstermann (2011) reported a percentage of residents being

Table 3 Results of ANCOVA-models analysing 1-year follow-up (n = 68). Dependent variable

p Value (model)

R2-coefficient of determination

QUALIDEM: total score at t2 p = .016 0.286 Non-significant independent variables: 1, 2, 3, 4, 5, 6, 8 (see underline) QUALIDEM: care relationship at t2 p = .023 0.273 Non-significant independent variables: 1, 2, 3, (4), 5, 6, 8 (see underline) QUALIDEM: positive affect at t2 p = .001 0.378 Non-significant independent variables: (1), 2, 3, 4, 5, 6, 8 (see underline) QUALIDEM negative affect at t2 p = .021 0.277 Non-significant independent variables: 1, (2), 3, 4, 5, 6, 8 (see underline) QUALIDEM restless tense behaviour at t2 p = .003 0.335 Non-significant independent variables: 1, 2, 3, 4, 5, 6, 8 (see underline) QUALIDEM positive self-image* at t2 p = .133 0.371 Non-significant independent variables: 1, 2, 3, 4, 5, 6, (7), 8 (see underline) QUALIDEM social relations at t2 p = .003 0.338 Non-significant independent variables: 1, 2, 3, 5, 6, 8 (see underline) QUALIDEM social isolation at t2 p = .109 0.212 Non-significant independent variables: 1, 2, 3, 4, 5, (6), 8 (see underline) QUALIDEM feeling at home* at t2 p = .348 0.283 Non-significant independent variables: 1, 2, 3, 4, 5, 6, (7), 8 (see underline) QUALIDEM having something to do* at t2 p = .115 0.383 Non-significant independent variables: 1, 2, 3, 4, 5, 6, (7), 8 (see underline)

Significant independent variables

p Value, partial Eta2-coefficient

QUALIDEM total score at t1

p = .001, Eta2 = 0.183

QUALIDEM care relationship at t1length of stay QUALIDEM positive affect at t1intervention/control group QUALIDEM negative affect at t1 sex QUALIDEM restless tense behaviour at t1

p p p p p p p

QUALIDEM positive self-image at t1

p = .090, Eta2 = 0.115

QUALIDEM social relations at t1length of stay QUALIDEM social isolation at t1at least one challenging behaviour at t2 QUALIDEM feeling at home* at t1

p p p p p

QUALIDEM having something to do* at t1

p = 0.055, Eta2 = 0.145

b .001, Eta2 = 0.239 = .094, Eta2 = 0.049 b .001, Eta2 = 0.286 = .091, Eta2 = 0.049 = .003, Eta2 = 0.141 = .079, Eta2 = 0.053 = .010, Eta2 = 0.112

b .001, Eta2 = 0.221 = 0.018,Eta2 = 0.094 = .016, Eta2 = 0.097 = 0.058,Eta2 = 0.062 = .077, Eta2 = 0.124

Independent variables: 1 = intervention/control group, 2 = sex, 3 = age, 4 = length of stay, 5 = stage of dementia (GDS) at t2, 6 = at least one challenging behaviour at t2, 7 = dependent variable at t1, 8 = QI-total-score; c.f. section “Methods”. *not applied for person with severe dementia (GDS = 7).

K. Wolf-Ostermann et al. / Applied Nursing Research 27 (2014) 33–40

underweight of 57.9% vs. 47.3% for PwD in SHA/special care units in nursing homes which are remarkably higher. Also rates of falls decreased and are lower than baseline results of Gulpers et al. (2011). Results of the quality development concerning QoL, activities of daily living and neuropsychiatric symptoms are discussed in detail below. Participants The distribution of male and female participants mirrors current trends in older adults and is comparable to those of Wolf-Ostermann et al. (2012) and te Boekhorst et al. (2009) and Verbeek et al. (2010) in the Netherlands. The mean age of participants is similar to findings in comparable studies in Germany (Wolf-Ostermann, et al., 2012; Wolf-Ostermann, et al., 2011). Compared to Wolf-Ostermann et al. (2012) similar results for ADL-functioning were achieved. Also the mean length of stay at baseline is comparable to similar studies in SHA (Wolf-Ostermann, et al., 2011). The mortality of participants during the 1-year study period was quite high but in line with results reported by van Dijk, Dippel, & Habbema (1991) for the 2-year mortality of PwD in ambulatory care and lower than results of a 1-year follow-up for newly admitted PwD in SHA in Germany (WolfOstermann, et al., 2012). Cognitive impairment The mean MMSE-score at baseline was 11.5, which corresponds to results in German SHA (Wolf-Ostermann, et al., 2012) or in group living homes in Sweden (Wimo, Adolfsson, & Sandman, 1995) and the Netherlands (te Boekhorst, et al., 2009; Verbeek, et al., 2010). During the 1-year follow-up a significant decline in both groups was found, but no significant differences between the groups. The decline in cognitive functioning reflects findings from similar studies (te Boekhorst, et al., 2009). Neuropsychiatric symptoms No significant changes in the occurrence of challenging behaviours were found. Verbeek et al. (2010) report similar results for agitation scores in small-scale living facilities remaining stable over time. For the defined ratio of residents with verbal agitation and the ratio of residents being verbally or physically aggressive a (not statistically significant) decline in the intervention group was seen for which at this point of time it could be only speculated that this may be the beginning of a positive effect of the quality development process. Quality of life In general, the assessed QoL of residents in SHA is moderate to high and comparable to the results of Wolf-Ostermann, et al. (2012). The nursing staff assessed very positively the QoL subscales “feeling at home” and “social relations”, which supports the assumption of SHA providing family-like structures for living. On the other hand, the subscale “having something to do” was assessed relatively low which is surprising thinking of the concept of SHA. One possible explanation for this may be that persons with a more severe dementia have very limited participation in household chores and inactivity is more striking in a family-like surrounding. The achieved results are also in line with results reported by te Boekhorst et al. (2009) and Verbeek et al. (2010) in small-scale living facilities/group living homes in the Netherlands. In the intervention group, a (statistically non-significant) increase of QoL was observed in some subdomains which may be pointing out that a guided quality development makes an impact, but such an interpretation can only be speculative. An explanation why the quality development process did not show statistically significant results in terms of better QoL might be that a biased sample in terms of SHA as a

39

whole was included into the study because only SHAs with already high quality standards were interested in participating. Another explanation might be the fact that the guided quality development itself lasted for 1 year and it might require a longer period of time before showing substantial results in psychosocial health outcomes and QoL. A lesson to be learned from this for future studies is that for the implementation and evaluation of quality development processes, long-term study intervals are essential. Limitations There are some possible limitations of this study that should be considered interpreting the results. One limitation is the relatively small number of participants and the short period of follow-up. Analysing a highly complex issue this may have the consequence of not being able to provide statistically evident results, particularly when taking into account that a quality development process might require some longer period of time before showing substantial results. Another limitation to be named is the fact that it was not possible to conduct an RCT randomizing all existing SHA but only those who showed a general interest in participating in this study. This may have led to the fact that in both groups highly engaged SHAs participated, which have already reached a high level of care and support for PwD and therefore cannot be improved easily to a greater extent. It cannot be ruled out that an exchange of information between the intervention and the control group might have occurred because of the physical proximity of all SHAs in Berlin. The consequence is that the comparison of quality levels between both groups can only be carried out with prudence. Furthermore, we were not able to obtain all QIs in the control group (e.g. resident–staff ratio). This was due to the fact that most of these SHA were not willing to provide this information. Further limitations could also be the different sources of information during data collection. Most of the assessments of the main outcome variables were done by specially trained RN. We therefore assume that quality of data and comparability are quite high. Concerning the measurement of QoL, it has to be noted that a proxy-measure was used which may differ substantially from a person's self-rating of QoL (Gräske, Fischer, Kuhlmey, & WolfOstermann, 2012). Nevertheless, a dementia-specific QoL questionnaire rated by professionals that can be applied in SHAs was used (Gräske et al., 2013). Conclusion The result of this study showed no convincing overall effects of an externally guided quality development process in SHAs for the primary outcome measures. No statistically significant difference in residents' QoL was found between intervention and control group. Nevertheless the results of our study contribute to literature in the way showing that a quality development process using tailored setting-specific QIs is feasible and the development of some of the proposed QIs over time showed promising beginning effects. We also want to emphasize the idea of proving empirical evidence for quality development processes. From a nursing practice perspective the results should raise the awareness of continuously ensuring quality of care and support for PwD by implementing tailored setting-specific quality development processes routinely even in settings which are not yet legally obligated to do this. This also implies that nurses are no longer in the role of just being responsible for performing quality management but change to an active role of developing and initializing quality management processes. Further research as well as longer study intervals is essential to demonstrate that setting-sensitive QIs are absolutely essential to measure quality of care in SHA. Quality development in SHA needs to be tailored to given conditions of structural and procedural features. It

40

K. Wolf-Ostermann et al. / Applied Nursing Research 27 (2014) 33–40

should be organized decentrally in order to show substantial results in terms of physical and psychosocial health outcomes of the residents. Acknowledgment The WGQual-study was funded by a grant from the German Federal Ministry of Education and Research. We express our gratitude to the co-operation partners who supported the study: “AnbieterVerbandqualitätsorientierterGesundheitspflegeeinrichtungen e.V. (AVG)“, „Caritas-verband für das Erzbistum Berlin e.V.“, “Verein zur Förderung altersgerechten Wohnens e.V. (FaW)“ and „Alzheimer-Gesellschaft Berlin e.V.“. We are indebted to all residents and nursing staffs who participated in our study. We gratefully thank Mr. Fidel Lim R.N., M.A, New York University, College of Nursing, for his helpful comments and proof-reading the paper. References Bouman, A. I. E., Ettema, T. P., Wetzels, R. B., van Beek, A. P. A., de Lange, J., & Dröes, R. M. (2011). Evaluation of Qualidem: A dementia-specific quality of life instrument for persons with dementia in residential settings; scalability and reliability of subscales in four Dutch field surveys. International Journal of Geriatric Psychiatry, 26(7), 711–722, http://dx.doi.org/10.1002/gps.2585. Cohen-Mansfield, J., Marx, M. S., & Rosenthal, A. S. (1989). A description of agitation in a nursing home. Journal of Gerontology, 44(3), M77–M84, http://dx.doi.org/ 10.1093/geronj/44.3.M77. Ettema, T. P., Dröes, R. -M., Lange, J., d., Mellenbergh, G. J., & Ribbe, M. W. (2007). QUALIDEM: Development and evaluation of a dementia specific quality of life instrument – Validation. International Journal of Geriatric Psychiatry, 22(5), 424–430, http://dx.doi.org/10.1002/gps.1692. Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). "Mini-mental state": A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189–198, http://dx.doi.org/10.1016/0022-3956(75) 90026-6. Gräske, J., Fischer, T., Kuhlmey, A., & Wolf-Ostermann, K. (2012). Quality of life in dementia care – Differences in quality of life measurements performed by residents with dementia and by nursing staff. Aging Ment Health, 16(7), 818–827, http:// dx.doi.org/10.1080/13607863.2012.667782. Gräske, J., Verbeek, H., Gellert, P., Fischer, T., Kuhlmey, A., & Wolf-Ostermann, K. (2013). How to measure quality of life in shared-housing arrangements? A comparison of dementia-specific instruments. Quality of Life Research, Epub first., http://dx.doi.org/ 10.1007/s11136-013-0504-8. Gräske, J., Wulff, I., Fischer, T., Meye, S., Worch, A., & Wolf-Ostermann, K. (2011). Ambulant betreute Wohngemeinschaften für ältere, pflegebedürftige Menschen – Unterstützung von Angehörigen und Ehrenamtlichen. Pflegezeitschrift, 64(11), 664–669. Gulpers, M. J. M., Bleijlevens, M. H. C., Ambergen, T., Capezuti, E., van Rossum, E., & Hamers, J. P. H. (2011). Belt restraint reduction in nursing homes: Effects of a multicomponent intervention program. Journal of the American Geriatrics Society, 59(11), 2029–2036, http://dx.doi.org/10.1111/j.1532-5415.2011.03662.x. Kane, R. A., Lum, T. Y., Cutler, L. J., Degenholtz, H. B., & Yu, T. C. (2007). Resident Outcomes in Small-House Nursing Homes: A Longitudinal Evaluation of the Initial Green House Program. Journal of the American Geriatrics Society, 55(6), 832–839, http://dx.doi.org/10.1111/j.1532-5415.2007.01169.x. Marquardt, G., & Schmieg, P. (2009). Dementia-friendly architecture: Environments that facilitate wayfinding in nursing homes. American Journal of

Alzheimer's Disease and Other Dementias, 24(4), 333–340, http://dx.doi.org/ 10.1177/ 1533317509334959. Meyer, S., Gräske, J., Worch, A., & Wolf-Ostermann, K. (2011). Ernährungssituation und Alltagsfähigkeiten bei Menschen mit und ohne Vorliegen einer Demenzerkrankung im Vergleich zwischen ambulant betreuten Wohngemeinschaften und Spezialwohnbereichen für Menschen mit Demenz. Praxis Klinische Verhaltensmedizin und Rehabilitation, 24(2), 117–125. Nakrem, S., Vinsnes, A. G., Harkless, G. E., Paulsen, B., & Seim, A. (2009). Nursing sensitive quality indicators for nursing home care: International review of literature, policy and practice. International Journal of Nursing Studies, 46, 848–857, http://dx.doi.org/10.1016/j.ijnurstu.2008.11.005. Prosiegel, M., Böttger, S., Schenk, T., König, N., Marolf, M., Vaney, C.,. Yassouridis, A. (1996). Der Erweiterte Barthel-Index (EBI) - eine neue Skala zur Erfassung von Fähigkeitsstörungen bei neurologischen Patienten [The Extended Barthel Index - a new scale for the assessment of disability in neurological patients.]. Neurologie & Rehabilitation, 1, 7–13. Reisberg, B., Ferris, S., de Leon, M., & Crook, T. (1982). The Global Deterioration Scale for assessment of primary degenerative dementia. The American Journal of Psychiatry, 139(9), 1136–1139. te Boekhorst, S., Depla, M. F. I. A., de Lange, J., Pot, A. M., & Eefsting, J. A. (2009). The effects of group living homes on older people with dementia: A comparison with traditional nursing home care. International Journal Of Geriatric Psychiatry, 24(9), 970–978, http://dx.doi.org/10.1002/gps.2205. van Dijk, P. T., Dippel, D. W., & Habbema, J. D. (1991). Survival of patients with dementia. Journal of the American Geriatric Society, 39(6), 603–610. Vasse, E., Moniz-Cook, E., OldeRikkert, M., Cantegreil, I., Charras, K., Dorenlot, P., & Vernooij-Dassen, M. (2012). The development of quality indicators to improve psychosocialcare in dementia. International Psychogeriatrics, 24(4), 921–930, http://dx.doi.org/10.1017/S1041610211002523. Verbeek, H., Zwakhalen, S. M. G., van Rossum, E., Ambergen, T., Kempen, G. I. J. M., & Hamers, J. P. H. (2010). Dementia care redesigned: Effects of small-scale living facilities on residents, their family caregivers, and staff. JAMDA, 11(9), 662–670, http://dx.doi.org/10.1016/j.jamda.2010.08.001. Wimo, A., Adolfsson, R., & Sandman, P. O. (1995). Care for demented patients in diffrent living conditions: Effects on cognitive function, ADL-capacity and behaviour. Scandinavian Journal of Primary Health Care, 13(3), 205–210, http://dx.doi.org/ 10.3109/02813439508996762. Wimo, A., & Prince, M. (2010). World Alzheimer Report 2010. Alzheimer´s Disease International. : The Global Economic Impact of Dementia. Wolf-Ostermann, K., & Fischer, T. (2010). Mit 80 in die Wohngemeinschaft – Berliner Studie zu Wohngemeinschaften für pflegebedürftige Menschen. Pflegewissenschaft, 2010(5), 261–272. Wolf-Ostermann, K., Worch, A., Fischer, T., Wulff, I., & Gräske, J. (2012). Health outcomes and quality of life of residents of shared-housing arrangements compared to residents of special care units - results of the Berlin DeWeGE-study. Journal of Clinical Nursing, 21(21–22), 3047–3060, http://dx.doi.org/10.1111/ j.1365-2702.2012.04305.x. Wolf-Ostermann, K., Worch, A., Meyer, S., & Gräske, J. (2013). Ambulant betreute Wohngemeinschaften für Menschen mit Pflegebedarf - Eine Studie zu gesetzlichen Rahmenbedingungen und Versorgungsangeboten in Deutschland. Zeitschrift für Gerontologie und Geriatrie, 1–7, http://dx.doi.org/10.1007/s00391-013-0524-6. Wolf-Ostermann, K., Worch, A., Wulff, I., & Gräske, J. (2011). Ambulant betreute Wohngemeinschaften für pflegebedürftige ältere Menschen – Angebots- und Nutzerstrukturen. Praxis Klinische Verhaltensmedizin und Rehabilitation, 24(2), 83–96. Worch, A., Gräske, J., Dierich, K., & Wolf-Ostermann, K. (2011). Wissensbasierte Qualitätsindikatoren zur Verbesserung gesundheitsbezogener Zielgrößen für Menschen mit Demenz in ambulant betreuten Wohngemeinschaften. Praxis Klinische Verhaltensmedizin und Rehabilitation., 24(2), 139–153. Wulff, I., Gräske, J., Fischer, T., & Wolf-Ostermann, K. (2011). Versorgungsstrukturen für ältere, pflegebedürftige Menschen mit und ohne Vorliegen einer Demenzerkrankung im Vergleich zwischen ambulant betreuten Wohngemeinschaften und Spezialwohnbereichen vollstationärer Einrichtungen. Praxis Klinische Verhaltensmedizin und Rehabilitation, 24(2), 97–105.

Quality of care and its impact on quality of life for care-dependent persons with dementia in shared-housing arrangements: results of the Berlin WGQual-study.

The aim of study was to evaluate setting-specific quality indicators (QIs) for shared-housing arrangements (SHA) regarding effects of a guided quality...
479KB Sizes 0 Downloads 0 Views