Haugan et al.

Health and Quality of Life Outcomes (2020) 18:1

https://doi.org/10.1186/s12955-019-1245-3

RESEARCH Open Access

Assessing quality of life in older adults:

psychometric properties of the OPQoL-brief
questionnaire in a nursing home
population
Gørill Haugan1,2* , Jorunn Drageset3,4, Beate André1, Kamile Kukulu5, James Mugisha6 and Britt Karin S. Utvær7

Abstract
Background: Well-adapted and validated quality-of-life measurement models for the nursing home population are
scarce. Therefore, the aim of this study was to test the psychometrical properties of the OPQoL-brief questionnaire
among cognitively intact nursing home residents. The research question addressed evidence related to the
dimensionality, reliability and construct validity, all of which considered interrelated measurement properties.
Methods: Cross-sectional data were collected during 2017–2018, in 27 nursing homes representing four different
Norwegian municipalities, located in Western and Mid-Norway. The total sample comprised 188 of 204 (92% response rate)
long-term nursing home residents who met the inclusion criteria: (1) municipality authority’s decision of long-term nursing
home care; (2) residential time 3 months or longer; (3) informed consent competency recognized by responsible doctor
and nurse; and (4) capable of being interviewed.
Results: Principal component analysis and confirmative factor analyses indicated a unidimensional solution.
Five of the original 13 items showed low reliability and validity; excluding these items revealed a good model
fit for the one-dimensional 8-items measurement model, showing good internal consistency and validity for
these 8 items.
Conclusion: Five out of the 13 original items were not high-quality indicators of quality-of-life showing low reliability and
validity in this nursing home population. Significant factor loadings, goodness-of-fit indices and significant correlations in
the expected directions with the selected constructs (anxiety, depression, self-transcendence, meaning-in-life, nurse-patient
interaction, and joy-of-life) supported the psychometric properties of the OPQoL-brief questionnaire. Exploring the essence
of quality-of-life when residing in a nursing home is highly warranted, followed by development and validation of new
tools assessing quality-of-life in this population. Such knowledge and well-adapted scales for the nursing home population
are beneficial and important for the further development of care quality in nursing homes, and consequently for quality-of-
life and wellbeing in this population.
Keywords: Factor analysis, Nursing home residents, Nursing home care, OPQoL-brief questionnaire, Psychometric
properties, Quality of life, Wellbeing

* Correspondence: [email protected]
1
NTNU Center for health promotion research, Norwegian University of
Science and Technology, Trondheim, Norway
2
Faculty of Nursing and Health Sciences, Nord University, Bodø, Norway
Full list of author information is available at the end of the article

© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 2 of 14

Background loneliness [5]; particularly when moving to a NH. With

Currently, the world faces a shift to an older population; advancing age, it is inevitable that people lose connec-
125 million people are now aged 80 years or older [1]. tion with their friendship networks and that they find it
While this shift started in high-income countries (for ex- more difficult to initiate new friendships and to belong
ample in Japan 30% of the population are already over 60 to new networks. However, a link between QoL and
years old), it is now low- and middle-income countries connectedness is emerging in the literature [5]. Despite
that are experiencing the greatest change. Today, most old age, chronical diseases or frailty; the desire for affili-
people can expect to live into their sixties and beyond [1]. ation and social bonding is an intrinsic human need, also
Between 2015 and 2050, the proportion of the world’s when living in a NH. Hence, the life situation for older
population over 60 years will nearly double from 12 to adults in NHs might differ significantly from other older
22%; by 2050, the world’s population aged 60 years and adult populations, staying at home or in hospitals. Con-
older is expected to total 2 billion, up from 900 million in sequently, a valid and reliable scale assessing QoL in this
2015 [1, 2]. All countries in the world face major chal- population is important for the further development of
lenges to ensure that their health and social systems are care quality and health promoting intervention in NHs.
ready to make the most of this demographic shift [1]. While planning for the present study, we searched,
As people live longer it is important to ensure that the broadly and thoroughly, for a valid and reliable measure
extra years of life are worth living, despite chronic ill- of QoL suitable for the NH population. Along this road,
nesses. Quality-of-life (QoL) and health promotive initia- we found the OPQoL-brief questionnaire [22], which
tives for older persons living in nursing homes (NH) will was developed by a “bottom-up” approach and tested
become ever more important in the years to come. The among older adults in Britain [22]. In a frail and vulner-
World Health Organization Quality of Life Group [3] able population such as the NH population, a shorter
defined QoL as an “individuals’ perception of their pos- scale is warranted. Though shorter instruments are more
ition in life in the context of the culture and value sys- limited in scope and sensitivity than longer measures,
tems in which they live and in relation to their goals, the benefits are reduced respondent and research burden
expectations, standards and concerns.” Additionally, glo- and costs. The OPQoL-brief is a shortened version of
bal QoL encompasses multiple constructs such as phys- the OPQOL-35 showing good psychometrics among
ical health, psychological status, independence level, older adults [20]; therefore, the OPQoL-brief was se-
social relationships, and relationship with significant fea- lected for this Norwegian study. To the authors know-
tures of the environment [3]. ledge, the OPQoL-brief has not previously been tested
QoL conceptual models and instruments for research, by means of confirmatory factor analysis as well as
evaluation and assessment in diverse populations have among NH residents.
been developed since the middle of last century [4, 5].
However, well-adapted and validated QoL measurement Aims
models for the NH population are scarce. Accordingly, The aim of this study was to assess the psychometric
our understanding of QoL among individuals living in properties of the Norwegian version of the OPQoL-brief
NHs is still limited. During the last decades, spirituality questionnaire in a cognitively intact (not diagnosed with
[6–8], a sense of meaning-in-life, hope, self-transcendence dementia and recognized by the responsible doctor and
[9–17], social relationships and social support [5] have be- nurse to have informed consent competency) NH popu-
come vital aspects of wellbeing and QoL among older lation. The research question was two-fold; (a) how well
adults. does the original one-factor measurement model of the
Global QoL is a multifaceted concept, representing a OPQoL-brief fit to the observed data? (b) Does the
subjective state which is affected by chronic and debili- OPQoL-brief reveal good reliability and construct valid-
tating health conditions [18]; such conditions are com- ity in a NH population? We expected the OPQoL- brief
mon among older adults in NHs. Research shows that to correlate with some established concepts, and tested
the NH population is characterized by high age, frailty, the following hypotheses:
mortality, disability, powerlessness, dependency, vulner-
ability, poor general health and a high symptom burden Hypotheses1 (H1): OPQoL-brief correlates negatively
[19–21]. Accordingly, moving to a NH results from nu- with anxiety and depression.
merous losses, illnesses, disabilities, loss of functions and Hypotheses2 (H2): OPQoL-brief correlates positively
social relations, and facing the end-of-life, all of which with self-transcendence, meaning-in-life, nurse-patient
detrimental to people’s functioning, independency and interaction and joy-of-life.
QoL. Moreover, older people experience changes in
roles, relationships, and living environments that can in- In accordance to the Standards for Educational and Psy-
crease their risk for experiencing social isolation and chological Testing [23, 24], the present research question
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 3 of 14

addressed evidence related to the dimensionality, reliability Norway. The total sample comprised 188 of 204 (92%
and construct validity, all of which considered interrelated response rate) long-term NH residents who met the in-
measurement properties. Dimensionality examines the ex- clusion criteria: (1) municipality authority’s decision of
tent to which the internal components of a test match the long-term NH care; (2) residential time 3 months or lon-
defined constructs, and is concerned with the homogeneity ger; (3) informed consent competency recognized by re-
of the items [25]. Reliability involves an instrument’s con- sponsible doctor and nurse; and (4) capable of taking
sistence and relative lack of error [25]. This study assessed adequately part in an interview situation. A nurse at the
internal consistence by the reliability coefficients Cron- actual ward presented potential participants with oral
bach’s alpha (α) and composite reliability (ρc). Construct and written information about the study, their rights as
validity refers to how well a scale actually measures the participants and their right to withdraw at any time.
construct it is intended to measure, and is based among Due to impaired vision, problems holding a pen, fa-
others on the constructs’ relationships to other variables tigue etc., this population have difficulties completing a
[25]. There are two subsets of construct validity: convergent questionnaire on their own. Therefore, six trained re-
construct validity and discriminant construct validity. Con- searchers (3 in each part of Norway) conducted one-on-
vergent construct validity tests the relationship between the one interviews in the resident’s private room in the NH.
construct and a similar measure; this shows that constructs Researchers with identical professional background (RN,
which are meant to be related are related. Discriminant MSc, trained and experienced in communication with
construct validity tests the relationships between the con- elderly, as well as teaching gerontology at an advanced
struct and an unrelated measure; this shows that the con- level) were trained to conduct the interviews in the same
structs are not related to something unexpected. In order manner. The OPQoL-brief was part of a battery of seven
to have good construct validity one must have a strong rela- scales comprising in total 120 items. To avoid misunder-
tionship with convergent construct validity and no relation- standings, interviewers read each question loudly, and
ship for discriminant construct validity [26]. In line with held a large-print copy of questions and possible re-
the WHO statement of health, salutogenic concepts such sponses in front of the participants.
as meaning, self-transcendence, joy-of-life and nurse-
patient-interaction are found to enhance NH residents’ Participants
QoL [10, 17, 20, 27–35], and to decrease anxiety and de- Participants ages ranged between 63 and 104 years
pression [36–38]. Therefore, these constructs were selected (mean 87.4 years, SD = 8.6). The sample consisted of 132
for assessing convergent construct validity by means of cor- women (73.3%) and 48 men (26.7%), where the mean
relational analyses. age for women was 88.3 years (SD = 1.8) and 86 years
Content validity refers to the degree to which a scale (SD = 1.2) for the men. In total, 23 were married, 22
has an appropriate, relevant sample of items to represent cohabitating, 1 was single, 106 were widows/widowers,
the construct of interest—that is, whether the content of and 37 were divorced.
the specific construct is adequately represented by the
items, meaning that the indicators measure all ideas in Instruments
the theoretical definition [39]. A frequent challenge oc- The measure of QoL analyzed here is the OPQOL-
curs when the wording of items is too similar—namely, brief – the short form of the OPQOL-35 questionnaire
the coefficient alpha, as well as the content validity and which was designed to assess QoL among older adults
dimensionality, are artificially enhanced. Nevertheless, 65+ [41, 42]. The OPQOL-35 has been validated on
items worded too similarly increase the average correl- community-dwelling older populations, and ethnically
ation among items, which in effect increases the coeffi- diverse population samples in Britain [43, 44]. The
cient alpha, yet without adding substantively to the OPQOL-35 was further tested among geriatric service
content validity of the measure. Although some similar- out-patients in Italy showing excellent applicability to
ity among items of a scale is needed to tap into the do- cognitively intact older people, and also to be applic-
main, several items that are mere rewordings of other able to most of the people suffering from mild or
items are redundant and contain very little new informa- moderate dementia [45–47]. The OPQOL-35 assumes
tion about the construct [40]. In that sense, theory, val- that QoL is a multidimensional concept; the original
idity, reliability, and dimensionality are intertwined. version includes eight domains [43, 44]. Nevertheless,
the factor structure has shown to be unclear; studies
Methods of the OPQOL-35 have reported two [43], four [43],
Design and data collection seven [48], or nine-factor solutions [42] based on prin-
Data were collected during 2017–2018 in 27 NHs repre- cipal component analysis (PCA). Like the original 35-
senting two small and one large urban municipality in items version [43], Chen [49] extracted eight factors
Mid-Norway and a large urban municipality in Western using PCA. No other factor analyses are currently
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 4 of 14

available. A more detailed examination of the factor The intention was to identify essential characteristics of
structure by means of CFA is needed [48]. NH patients’ experiences of JoL in their daily life. The
The OPQoL-brief comprises of 13 items which are JoLS covers domains that identify fundamental qualities
scored Strongly agree = 1, Agree = 2, Neither = 3, Dis- stressed in well-being theory [69–72], nursing care lit-
agree = 4, Strongly disagree = 5 [22]. The items are erature [73–76], and the dimensions found by 29 in-
summed for a total OPQoL-brief score, then positive depth interviews on the essence of joy-of-life with NH
items are reverse coded, so that higher scores represent residents [77]. A 13-items version of the JoLS was found
higher QoL. The total sum-score ranges from 13 to 65. reliable and valid in this population (Haugan, Rinnan
Examples of items include enjoying one’s life, looking et al.2019), and was applied in the present study.
forward to things, staying involved with things, and feel- The Nurse-Patient-Interaction Scale (NPIS) developed
ing safe where one lives, etc. (Table 4 in Appendix 1). in Norway, assessed perceived nurse-patient-interaction.
The OPQoL-brief was found to be a highly reliable and The NPIS comprises 14 items identifying essential rela-
valid measure of QoL in old age [22]. For use in this tional qualities stressed in the nursing literature [31].
study, two experts of both languages, English and Nor- This scale is scored from 1 (not at all) to 10 (very much);
wegian, translated the OPQoL-brief into Norwegian, fol- total score ranges between 14 and 140, where higher
lowing the procedure of back-and-forth translation. To numbers indicate better perceived nurse-patient-
better reflect the nuances of the target language [50], interaction. The NPIS has shown good psychometric
two independent translators did the forward translation properties with good content validity and reliability
into Norwegian (their mother tongue) [51]. One of these among NH residents [31].
was a naive translator who was unaware of the objective The Hospital Anxiety and Depression Scale (HADS),
of the questionnaire, while the other was a researcher in comprising 14 items, with subscales for anxiety (HADS-
the field of QoL. No discrepancies appeared. To assure A, 7 items) and depression (HADS-D, 7 items) assessed
the accuracy of the translation, the initial translation was anxiety and depression. Each item is rated from 0 to 3,
independently back-translated (from Norwegian into giving a range of total score between 0 and 21; higher
English) by two independent translators. The back- scores indicate more anxiety and depression. The HADS
translators were not aware of the intended concept the has shown good to acceptable reliability and validity in
questionnaire was [52]. the NH population [78].
The Self-Transcendence Scale (STS) [53] assessed inter-
personal and intrapersonal self-transcendence. The STS Ethical considerations
comprises 15 items, each with a score of 1–4, reflecting We obtained approval by the Regional Committee for
expanded boundaries of self which are considered to be Medical and Health Research Ethics in Norway (ref.nr
characteristics of a matured view of life [54]. Total score 2014/2000/REK Central) as well as from the Manage-
ranges between 15 and 60, where higher scores indicate ment Units at the 27 NHs. Each participant provided
higher ST. The STS has shown good psychometric prop- voluntarily written informed consent.
erties [55, 56] and has been translated into Norwegian,
and validated in NH patients [56] showing a two-factor- Data analysis
construct (STS1 & STS2) to be most valid and reliable The same data were analyzed by descriptive statistics
[54]. The present study applied this two-factor construct and principal component analysis (PCA) using IBM
(ST1, ST2). SPSS version 25, and confirmatory factor analysis (CFA)
The Purpose-in-Life Test (PIL) assessed meaning-in-life. by means of Stata 15.1 [79]. When evaluating a measure-
Based on Frankl’s theory, the PIL was designed to be a ment scale investigating the underlying dimensionality
general tool assessing meaning [57–60] and has been of data and the adequacy of each individual item is cen-
commonly used for this purpose [61–63]. The PIL is tral. In these instances, PCA and CFA can provide com-
translated into Norwegian [64] and has previously been plementary perspectives on data, giving different pieces
used with elderly individuals up to 104 years old [65–67]. of information [25, 80]. The implicit assumption under-
The Norwegian version has been validated among NH lying the use of PCA in the present study is the insecur-
residents, showing good psychometric properties [66]. ity with respect to the dimensionality of the OPQoL-
Each statement is scored from 1 to 7; four represents a Brief, which has not been previously tested by means of
neutral value, whereas the numbers from 1 to 7 stretch CFA, neither among NH residents. As previously pre-
along a continuum from one extreme feeling to the op- sented, the OPQoL-brief is a short version of the original
posite kind of feeling; higher scores reflect higher OPQoL scale, which has shown 2,4,7,8 and 9 factors.
meaning-in-life [60]. Total score ranges from 20 to 140. Therefore, a broad perspective on the observed data
The Joy-of-Life scale (JoLS) was developed in Norway using PCA followed by the confirmation procedure was
to assess NH patients’ perceived joy-of-life (JoL) [68]. used.
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 5 of 14

Confirmatory factor analysis (CFA) is a sub-model in this study, 11 (6.1%) of the NH residents reported QoL
structural equation modeling that deals specifically with mean-score < 3.0 interpreted as a low QoL, 90 (49.7%)
measurement models [81], accounting for random meas- stated a high QoL ≥4.0–5.0, while 77 (42.5%) reported
urement error, and thus derive a more accurate evaluation QoL mean-scores between 3.0–3.99, representing a
of the psychometric properties of the scales used. A high modest QoL. Figure 1 displays the distribution of the
loading of an item indicates that there is much in com- OPQoL mean-scores, while Table 1 lists the means,
mon between the factor and the respective item [82]. standard deviation, Cronbach’s alpha and correlation
Loadings below 0.32 are considered poor, ≥0.45 fair, ≥0.55 matrix for the constructs of QoL, ST1, ST2, PIL, JoL,
good, ≥0.63 very good, and above 0.71 are excellent [82]. NPIS, HADS-A and HADS-D.
Thus, a good rule of thumb for the minimum loading is
.32 [83], which equates to approximately 10% overlapping
variance with the other items in the factor. A “cross-load- Dimensionality
ing” item loads at .32 or higher on two or more factors. Principal Component Analysis (PCA)
The present study assessed model fit adequacy by χ2- In order to explain as much of the total variance as
statistics and various fit indices. In line with the ‘rules of possible with as few factors as possible, the OPQoL-
thumb’ given as conventional cut-off criteria [84] the fol- brief was subjected to PCA. The Kaiser-Meyer-Olkin
lowing fit indices were used; χ2-statistics, the Root Mean measure of sampling adequacy exceeded the recom-
Square Error of Approximation (RMSEA) and the Stan- mended value of .60 (.84) and Bartlett’s test of spher-
dardized Root Mean Square Residual (SRMS) with icity showed statistical significance (p < 0.0001),
values below 0.05 indicating good fit, whereas values supporting the factorability of the correlation matrix.
smaller than 0.10 is interpreted as acceptable [85]. Fur- We search for the cleanest structure of the concept
ther, the Comparative Fit Index (CFI) and the Tucker- under investigation and expected the OPQoL-brief to
Lewis Index (TLI) with acceptable fit set at 0.90 [84, 86] be one- or multi-dimensional with correlated factors.
were used. Both skewness and kurtosis were significant Hence, an oblique rotation such as promax should
and the Robust Maximum Likelihood (RML) estimate theoretically render a more accurate solution [89].
procedure was applied. When analyzing continuous but PCA with promax rotation and Kaiser Normalization
non-normal endogenous variables, the Satorra-Bentler were used; three factors with eigenvalue 1.0 and
corrected χ2 [87] should be reported [88]. greater (4.82, 1.72 and 1.06, respectively) were ex-
tracted (Table 2). Figure 2 portrays the scree-test of
Results the OPQoL-brief data showing the number of factors
Descriptives to retain is three. Table 2 lists the loadings and
The OPQoL-brief 13-items mean-scores ranged between variance for this rotated 3-factor solution of the
2.99–4.53, showing a total mean of 3.9 (SD = 1.01). In OPQoL-brief suggested by PCA. Yet, this 3-factor

Fig. 1 Histogram: The distribution of the OPQoL mean-score, 13-items

Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 6 of 14

Table 1 Distribution of the OPQoL scores, Means (M), Standard Confirmatory Factor Analysis (CFA)
deviations (SD), Cronbach’s alpha, Correlation coefficients for Firstly, we checked the original 13-items unidimensional
OPQoL to Self-Transcendence, Meaning-in-life, Sense of version, revealing a very bad fit to the present data. Con-
Coherence, Nurse-patient Interaction, Joy-of-Life, Depression, sequently, we tested the 3-factor solution suggested by
and Anxiety the PCA (Factor 1: items 3, 7, 8, 10, 11, 12; Factor 2:
Distribution of the OPQoL scores items 2, 5, 6, 7 and Factor 3: items 1, 4, 9, 13). Running
OPQoL score 1–2.99 3.0–3.99 4.0–5.0 CFA, this 3-factor-model did not fully converge and did
N = 181 not provide fit indices; both of which indicating misspe-
100% 11 (6.1%) 77 (42.5%) 90 (49.7%) cifications. The original OPQoL-brief revealed one di-
Variable Cronbach’s Mean Std.Dev. Correlations (r2) mension; and a 3-factor-solution of the OPQoL-brief
(number items) Alpha (α) (M) (SD) OPQoL brief (13) construct did not seem theoretically meaningful. There-
OPQoL-brief 0.83 3.901 1.008 1.00 fore, we turned back to the original unidimensional 13-
(13)
items model [22] for further examination.
ST1 (7) 0.65 2.518 0.569 0.56**
ST2 (8) 0.68 3.145 0.449 0.47**
PIL (20) 0.80 3.482 1.091 0.40** Reliability
NPIS (14) 0.90 7.981 1.907 0.45** Model-1 – the original OPQoL-brief unidimensional version
Model-1 comprising 13 items gave significant t-values
JOL (13) 0.88 5.682 1.137 0.69**
for all estimates, showing completely standardized factor
HADS-D (7) 0.74 1.686 0.552 − 0.17**
loadings from .78–.41, and squared multiple correlations
HADS-A (7) 0.83 1.864 0.410 − 0.29** (R2) ranging between .61–.16. Some items (item4,5,6,13)
OPQoL-brief Quality-of-life, ST1 Interpersonal self-transcendence, ST2 disclosed low R2-values (≤0.19) indicating low reliability.
Intrapersonal self-transcendence, PIL Purpose in Life test, NPIS Nurse-Patient
Interaction, JoL Joy-of-Life Scale, HADS-D Depression, HADS-A Anxiety, N = 181
The model fit was bad: χ2 = 236.36, (df = 65), χ2/df =
3.64, p = 0.0001, RMSEA = 0.12, p-value for test of close
solution revealed 9 cross-loadings, with substantial fit = 0.0001, CFI = 0.75, TLI = 0.70, and SRMR = 0.094.
factor loadings on all factors indicating an unclear However, composite reliability for this one-factor con-
dimensionality. struct was good (ρc = 0.84), indicating good reliability
Substantial conclusions based solely on PCA should (values ≥0.6 is considered acceptable, while values ≥0.7
not be drawn [89]; therefore, we turned to confirmatory are good) [84, 90]. The alpha levels for the various mea-
factor analysis (CFA). sures indicated an acceptable inter-item consistency with

Table 2 Exploratory Factor Analysis of the OPQoL-brief questionnaire – Rotated Component Matrix. Estimates for factor loadings,
extraction sums of squared loadings and Cronbach’s alpha
Model-1 (3 factors, 13 items)
OPQoL1 I enjoy my life overall .489 .565 .706
OPQoL2 I look forward to things .713 .632
OPQoL3 I am healthy enough to get out and about .505 .367 .501
OPQoL4 My family, friends or neighbours will help me if needed .569 .634
OPQoL5 I have social or leisure activities/hobbies that I enjoy doing .807
OPQoL6 I try to stay involved with things .800
OPQoL7 I am healthy enough to have my independence .701
OPQoL8 I can please myself what I do .576 .520 .455
OPQoL9 I feel safe where I live .525 .617
OPQoL10 I get pleasure from my home .719 .347
OPQoL11 I take life as it comes and make the best of things .601 .591
OPQoL12 I feel lucky compared to most people .690 .340
OPQoL13 I have enough money to pay for household bills .684
Cumulative % of total variance explained 34.967 48.199 56.326
Cronbach’s Alpha (number of items) .75 (6) .66 (3) .62 (4)
Extraction Method: Principal Component Analysis. Rotation Method: Promax with Kaiser Normalization. Values< 0.32 are suppressed. Model-1: Three components
extracted based on Eigenvalue > 1. Total variance explained: 54,325. Rotation converged in 8 iterations. Listwise N = 181
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 7 of 14

Fig. 2 Scree-plot of the OPQoL Brief questionnaire, 13-items

Cronbach’s alpha coefficients of 0.65–0.90 (Table 1) and Therefore, it is theoretically rational that they revealed a
composite reliability of 0.80–0.84 (Table 3). very high MI. Item3 loaded higher than item7; for that
reason, item7 was dismissed from the model. The next
Construct validity step was to consider item6 (‘I try to stay involved with
An inspection of the standardized residuals and the things’), which displayed an extremely high MI with
modification indices (MIs), discovered five significant re- item5 (‘I have social or leisure activities/hobbies that I
siduals [item7–3, (0.31) item6–4 (0.21), item5–6 (0.49), enjoy doing’). Item6 loaded significantly lower than item5
item 11–5(− 0.21), item11–6 (− 0.20)]. Furthermore, ten and was dismissed, and the model was run once more.
pair of items showed MIs higher than 10, all of which This 11-items version gave somewhat better fit (χ2 =
pointing to misspecifications. For the pairs of items 3–7 137.62, (df = 44), χ2/df = 3.13, p = 0.0001, RMSEA = 0.11,
and items 5–6 the MIs were extremely high (MI = 15.10 p-value for test of close fit = 0.0001, CFI = 0.83, TLI =
and MI = 36.53, respectively). 0.78, SRMR = 0.080), although, a poor fit. Now, only one
Item3 (‘I am healthy enough to get out and about’) and residual was significant, involving the pair of item11–5.
item7 (‘I am healthy enough to have my independence’) Still, several very high MIs were found, involving item11
contain physical functioning and thus share variance. (‘I take life as it comes and make the best of things’),

Table 3 Goodness-of-fit measures for OPQoL-brief measurement model. Confirmatory Factor Analysis for Model-1, Model-2 and
Model-3
Fit Measure Model-1 N = 181 13 items Model-2 N = 181 9 items Model-3 N = 181 8 items
χ2 Satorra Bentler 236.358 54.213 31.547
p-value 0.00001 0.001 0.048
χ
2 1
3.33 (Df = 65) 2.008 (Df = 27) 1.58 (Df = 20)
df Satorra Bentler
RMSEA 0.121 (CI: 0.104–0.137) 0.074 (CI: 0.045–0.103 0.056(CI: 0.005–0.092)
p-value (close fit test) 0.000001 0.080 0.359
SRMR 0.094 0.060 0.050
CFI 0.76 0.93 0.97
TLI 0.70 0.91 0.95
Average Variance extracted (AVE) 0.300 0.324 0.340
ðΣλÞ2 0.84 0.80 0.80
pc ¼ ½ðΣλÞ2 þΣðθÞ

OPQoL Quality of Life measurement model. RMSEA Root Mean Square Error of Approximation. SRMS Standardized Root Mean Square Residual, CFI The
Comparative Fit Index, TLI Tucker-Lewis Index, 1Df Degrees of freedom, ρc Composite reliability. Model-1: 13 items, Model-2: 9 items (items 6, 7, 10, 11 are
dismissed). Model-3: 8 items (items 6, 7, 10, 11 and 12 are dismissed). Listwise N = 181
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 8 of 14

indicating this item to share unexplained variance with a the OPQoL-brief revealed good reliability and construct
number of other items. Hence, item 11 was excluded. validity in a Norwegian NH population. Thus, the re-
Next, the estimates pointed at item10 (‘I get pleasure search question addressed evidence related to the di-
from my home’) and item9 (‘I feel safe where I live’); both mensionality, reliability and construct validity of the
concerned with resident’s sense of home while residing OPQoL-brief questionnaire in this population.
in a NH. Item9 showed the best loading and was kept,
while item10 was set aside. Dimensionality
The scree-test portrayed in Fig. 2 indicated that the
Model-2 – the OPQoL-brief 9-items unidimensional version number of factors to retain was three. However, two fac-
This modified version (including items 1,2,3,4,5,8,9,12,13), tors showed eigenvalues substantially higher than one,
framed Model-2, gave an acceptable fit (χ2 = 54.21, (df = 27), while the third factor was close to one (1.06), along with
χ2/df = 2.01, p = 0.001, RMSEA = 0.074, p-value for test of the next factors showing eigenvalues of 0.98, and 0.88,
close fit = 0.080, CFI = 0.93, TLI = 0.91, SRMR = 0.06). How- respectively. Hence, it seems not reasonable to regard
ever, even not a good fit. the third factor with eigenvalue of 1.06 as ‘major’ and
the fourth with eigenvalue of 0.98 as ‘trivial’. When it
Model-3 – the OPQoL-brief 8-items unidimensional version. comes to determining the number of factors, Kaiser’s
Finally, dismissing item12 (‘I feel lucky compared to most method (K1) sometimes is problematic and inefficient
people’) gave a god fit to the present data: χ2 = 31.55, [91]. As seems to be the case here, the Kaiser-Guttman
(df = 20), χ2/df = 1.58, p = 0.048, RMSEA = 0.056, p-value rule of retaining eigenvalues larger than 1 is not inter-
for test of close fit = 0.359, CFI = 0.97, TLI = 0.95 and pretively useful because it tends to result in the retention
SRMR = 0.05. This version of the one-dimensional of too many factors [92]. Despite K1’s widespread use,
model including eight items (item 1,2,3,4,5,7,9 and 13) experts agree that it has deficiencies and that its use is
was framed Model-3, representing the best fitting model. not recommended [92]. PCA needs to balance parsi-
Figure 3 portrays Model-3, showing the factor loadings, mony with adequately representing underlying correla-
multiple squared correlations (R2), model fit and com- tions, so its utility depends on being able to differentiate
posite reliability (ρc). major factors from minor ones [91]. By looking at the
scree-plot for the PCA in the present study this issue
Discussion seems evident; one strong factor along with several small
When evaluating a measurement scale, researchers face factors were portrayed.
two important questions: (1) the underlying dimension- Moreover, the rotated 3-factor solution suggested by
ality of data (not too many, not too few factors), and (2) PCA revealed several cross-loadings with substantial
the adequacy of the individual items. This study assessed factor loadings on all factors, thwarting the dimension-
how well the original one-factor measurement model of ality. Only four (items 5,6,7,13) loaded solely on one di-
the OPQoL-brief fit to the observed data, and whether mension, indicating an unclear dimensionality of the

Fig. 3 OPQoL-brief measurement model including 8 items (1,2,3,4,5,8,9,13). N = 181. Standardized factor loadings, multiple squared correlations,
and composite reliability
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 9 of 14

construct and probably a one-dimensional solution like Content validity is a sub-form of construct validity, re-
Bowling et al. [22] presented. Turning to CFA, the ana- ferring to whether the OPQoL-brief has an appropriate,
lyses suggested a unidimensional solution (Table 3). relevant sample of items to represent the QoL construct.
However, some items seemed troublesome, indicating If the wording of items is too similar, a challenge occurs;
misspecifications. items worded too similarly increase the average correl-
ation among items, which in effect increases the coeffi-
Reliability cient alpha, yet without adding substantively to the
Reliability and construct validity are related to the ad- content validity of the measure. Firstly, items 3 and 7
equacy of the individual items; highly significant stan- possibly are worded too closely; ‘I am healthy enough to
dardized factor loadings–preferably > 0.7 indicates that get out and about’ (item3) and ‘I am healthy enough to
the items perform as good indicators for the QoL con- have my independence’ seem to measure the same as-
struct in the NH population. The square of a standard- pect. Staying in a NH without having dementia means
ized factor loading (R2), termed the variance extracted of that you on average have 6–7 diagnoses of chronic con-
the item, represents how much variation in an item the ditions [100], which negatively affect health, functioning
latent construct explains [93]. Loadings falling below 0.7 and independency. Largely, cognitively intact NH resi-
can still be significant, but more of the variance in the dents are not healthy enough to get out and about. Due
measure is error variance than explained variance. Look- to illness and health problems, followed by care needs,
ing at the factor loadings and the R2-values, only three they have moved to a NH. Consequently, their indepen-
items loaded good-excellent; item1 (λ = 0.82) was excel- dency is impeded; many NH residents perceive their
lent, while item2 (λ = 0.69) and item8 (λ = 0.66) dis- institutionalization as the beginning of their loss of inde-
played good loadings. Contrasting this, item4 (λ = 0.41, pendence and autonomy [101–103].
R2 = 0.17), item6 (λ = 0.42, R2 = 0.18), and item13 (λ = The notion that ‘I am healthy enough to have my inde-
0.44, R2 = 0.19) performed like invalid indicators of QoL; pendence…to get out and about’ might not indicate inde-
the OPQoL-construct explained only a limited amount pendence in the NH life situation very well; in fact, it
of the variance in these items. Consequently, the reliabil- could be the opposite. Striving for independence while
ity of these indicators was low. The other seven items you are totally dependent on others might damage your
displayed fair factor loadings ranging between .46–.57. QoL. Although some similarity among items of a scale is
Hence, reliability was acceptable, but not fully sup- needed to tap into the domain, several items that are
ported. An examination of the inter-item correlations re- mere rewordings of other items are redundant and con-
vealed plausible correlations (Table 5 in Appendix 2), tain very little new information about the construct [40].
with the highest values for the pair of items1–2 (r = .62), Secondly, the items concerning one’s home (items 9,
items5–6 (r = .55), items1–8 (r = .55) and items3–7 10), which for these older adults is a NH, might not be
(r = .47). Moreover, Cronbach’s alpha (α) (Table 1) and worded specific or precise enough. Many older adults in
composite reliability (ρc) (Table 3) revealed good values, NHs do not experience the NH as their home [102], and
indicating good internal consistency [84, 90]. are grieving over that they had to leave their home,
representing a loss to them. The NH is the last stop in
Construct validity their life. The expression that ‘I get pleasure from my
Construct validity deals with the accuracy of measurement, home’ (item10) might not be as central as it would be if
reflecting the extent to which a set of measured indicators these individuals were staying in their private home.
actually reflect the theoretical latent construct the items are However, ‘I feel safe where I live’ (item9) seems more ap-
designed to measure [94]. In the present study, convergent propriate; NH residents highlight the importance of feel-
construct validity was supported by significant negative cor- ing safe to their thriving and QoL [104, 105]. This
relations between OPQoL-brief and HADS-A and HADS- population is characterized by high age, numerous
D as well as positive correlations with ST1, ST2, PIL, NPIS losses, frailty, mortality, disability, powerlessness, de-
and JOL (Table 1). Both hypotheses (H1 and H2) were sup- pendency, vulnerability, poor general health, a high
ported. Items 1,2 and 8 revealed the best loadings, repre- symptom burden and facing the end-of-life [19–21], all
senting good indicators for QoL in the NH population. of which increases distress and vulnerability. Thus, feel-
Interestingly, item8 (‘I can please myself what I do’) loaded ing safe while staying in a NH seems closely connected
strongly (.66), implying to be a valid indicator of QoL in to the nurse-patient relationship, care quality and nurse-
this population. Considering that NH residents commonly patient interaction, more than being at ‘my home’.
experience idleness, spending many hours doing nothing, Hence, indicators including the NH working culture, mi-
waiting, sleeping, this finding is noteworthy. Doing some- lieu, atmosphere and nurse-patient interaction might be
thing, being active with something which you like, is essen- essential domains to include in a QoL measurement for
tial for QoL among NH residents [95–99]. NH residents [102]. Looking at the correlations between
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 10 of 14

the summative scores (Table 1), QoL correlated highly help fill in the questionnaires might have introduced
with the nurse-patient-interaction, along with joy-of-life, some bias on the respondents’ responses, which is a
interpersonal and intrapersonal self-transcendence and limitation of this study.
meaning-in-life. QoL correlated negatively with anxiety
and depression. Thus, convergent construct validity was
well supported. Conclusion
Item12 (‘I feel lucky compared to most people’) did not This study suggests a unidimensional solution of the
explain a substantial amount of the variation in the OPQoL-brief. However, five of the original 13 items
OPQoL-construct (R2 = 0.24). It might be difficult to emerged to be poor indicators of the OPQoL-construct
know who one should compare oneself with. If compar- showing fair reliability and an insufficient validity. The
ing with the healthy ones coping at home, one might not present study suggests that the nine- and eight-items
feel very lucky. Contrary, compared to those who are in versions revealed an acceptable and a good fit to the
hospital waiting for a place in a NH, one might feel data, respectively. Further development and testing of a
lucky. Probably this indicator could be more specified well-adapted scale assessing QoL in the NH population
towards the life situation of residing in a NH. are required.
Finally, item11 (‘I take life as it comes and make the
best of things’), revealed significant correlations with
many items involved in the OPQoL-brief questionnaire.
Possibly, item11 covers an attitude and coping mechan-
Appendix 1
ism which is very much needed and therefore commonly
developed among NH residents. Consequently, this indi-
Table 4 The OPQoL-brief questionnaire. Original 13-items ver-
cator largely relates with the other indicators, sharing
sion, plus the preliminary single item on global QoL (OPQoL-G).
variance, and thus blurring the dimensionality and the Means and Standard deviation. During the last week, to what
statistical fit. Including correlated error terms concern- extent have you experienced that...
ing item11 might be an option.
Variable N Mean Std.Dev.
OPQoL1 I enjoy my life overall 181 3.63 1.082
Limitations
The shortened OPQoL-brief construct was supported by OPQoL2 I look forward to things 181 3.89 0.958
significant factor loadings, several goodness-of-fit indices OPQoL3 I am healthy enough to get out and 181 3.48 1.269
about
and significant correlations in the expected directions
with the selected constructs. However, a good model fit OPQoL4 My family, friends or neighbours will 181 4.27 1.014
help me if needed
does not guarantee that we have obtained ‘the true
model’; other alternative models might fit the data OPQoL5 I have social or leisure activities/hobbies 181 2.99 1.251
that I enjoy doing
equally well as the model found [106].
OPQoL6 I try to stay involved with things 181 3.50 1.193
The effective (listwise) sample size was N = 181, which
is considered medium, and close to what is understood OPQoL7 I am healthy enough to have my 181 3.75 1.212
independence
as a large sample size. A rate of 10 cases per observed
variable is given as a rule of thumb [81, 90]. The models OPQoL8 I can please myself what I do 181 3.75 0.907
tested in this study included 13 items; accordingly, the OPQoL9 I feel safe where I live 181 4.53 0.764
sample of N = 181 should be enough. Out of 204 NH pa- OPQoL10 I get pleasure from my home 181 4.01 1.000
tients fulfilling the inclusion criteria, 188 participated, OPQoL11 I take life as it comes and make the 181 4.42 0.754
giving a response rate of 92%. This along with almost no best of things
missing data represent a strength of this study. OPQoL12 I feel lucky compared to most people 181 4.07 0.864
The OPQoL-brief scale was part of a questionnaire OPQoL13 I have enough money to pay for 181 4.42 0.835
comprising 120 items. Accordingly, frail older NH resi- household bills
dents might tire when completing the questionnaire, Items 6, 7, 10, 11 and 12 are omitted in the best fitting 8-items
representing a possible bias to their reporting. To avoid measurement model
Listwise N = 181. The OPQoL-brief is scaled 1–5, where higher score means
such a bias, we carefully selected and trained experi- higher QoL
enced researchers in conducting the interviews following
a standardized procedure, including taking small breaks
at specific points during the process. This procedure
worked out very well; all participants fulfilled the ques-
tionnaire without considerably difficulties. The fact that
the researchers visited the participants in the NHs to
Appendix 2
Table 5 Inter-item correlation matrix
OPQoL item 1I 2 I look 3 I am 4 My family, 5 I have social or 6 I try to 7 I am healthy 8 I can 9 I feel 10 I get 11 I take life as 12 I feel 13 I have
enjoy forward healthy friends or leisure activities/ stay enough to please safe pleasure it comes and lucky enough
my life to enough to neighbors will hobbies that I involved have my myself where from my make the best compared money to pay
overall things get out and help me if enjoy doing with independence what I I live home of things to most for household
about needed things do people bills
1 I enjoy my life 1,000
overall
2 I look forward ,622 1,000
to things
3 I am healthy ,447 ,399 1,000
enough to get
out and about
4 My family, ,345 ,409 ,200 1,000
Haugan et al. Health and Quality of Life Outcomes

friends or
neighbors will
help me if
needed
5 I have social or ,395 ,460 ,226 ,314 1,000
(2020) 18:1

leisure activities/
hobbies that I
enjoy doing
6 I try to stay ,269 ,422 ,206 ,348 ,548 1,000
involved with
things
7 I am healthy ,281 ,187 ,465 ,046 ,224 ,136 1,000
enough to have
my
independence
8 I can please ,547 ,412 ,353 ,241 ,363 ,314 ,341 1,000
myself what I do
9 I feel safe ,343 ,300 ,254 ,228 ,119 ,030 ,296 ,235 1,000
where I live
10 I get pleasure ,349 ,320 ,271 ,103 ,239 ,226 ,355 ,379 ,481 1,000
from my home
11 I take life as it ,426 ,206 ,304 ,104 ,025 -,017 ,311 ,383 ,426 ,310 1,000
comes and make
the best of things
12 I feel lucky ,357 ,246 ,216 ,032 ,220 ,133 ,285 ,345 ,304 ,442 ,375 1,000
compared to
most people
13 I have enough ,327 ,281 ,294 ,253 ,082 ,126 ,240 ,224 ,300 ,209 ,310 ,320 1,000
money to pay for
household bills
Page 11 of 14
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 12 of 14

Abbreviations 2. Kinsella K, He W. An Aging World: 2008, U.C. bureau, editor. Washington,

CFI: Comparative Fit Index; HADS: Hospital Anxiety and Depression Scale; DC: U.S. Department of Health and Human Services National Institutes of
HADS-A: Hospital Anxiety and Depression Scale, subscale assessing anxiety; Health NATIONAL INSTITUTE ON AGING U.S. Department of Commerce
HADS-D: Hospital Anxiety and Depression scale, subscale assessing economics and statistics; 2009. p 1-191.
depression; JOL: Joy-of-life; JoLS: Joy-of-life scale; MSc: Master of Science; 3. The WHOQOL Group, W., Development of the World Health Organization
N: Sample size; NH: Nursing home; NPIS: Nurse-Patient-Interaction Scale; WHOQOL-BREF quality of life assessment. The WHOQOL Group. Psychol
PIL: Purpose-in-life test; QoL: Quality of life; RMSEA: Root Mean Square Error Med., 1998;28(3): p. 551–8.
of Approximation; RN: Registered nurse; SD: Standard Deviation; 4. Bowling, A., Measuring health: a review of quality of life measurement
SEM: Structural Equation Modelling; SRMS: Standardized Root Mean Square scales. third ed. 2004, Maidenhead, GB. : Open University. 224.
Residual; ST: Self-Transcendence; ST1: Inter-personal Self-Transcendence; 5. Cooney A, et al. Exploration of the meaning of connectedness for older
ST2: Intra-personal Self-Transcendence; STS: Self-Transcendence Scale; people in long-term care in context of their quality of life: a review and
TLI: Tucker Lewis Index; WHO: The World Health Organization commentary. Int J Older People Nursing. 2014;9(3):192–9.
6. Bredle JM, et al. Spiritual well-being as a component of health-related
Acknowledgements quality of life: the functional assessment of chronic illness therapy—spiritual
The authors acknowledge the nursing home residents who voluntarily well-being scale (FACIT-Sp). Religions. 2011;2(1):77–94.
participated in this study, and the research assistants Eva Rinnan, Brit Moene 7. Chirico F. Spiritual well-being in the 21st century: it is time to review the
Kuven, Wenche Mjanger Eide, and Siw Eriksen Taasen who collected the current WHO’s health definition. J Health Soc Sci. 2016;1(1):11–6.
data. 8. Dhar N, Chaturvedi S, Nandan D. Spiritual Health Scale 2011: Defining and
Measuring 4th Dimension of Health. Indian J Community Med. 2011;36(4):275–82.
Authors’ contributions 9. Bradshaw SA, Playford ED, Riazi A. Living well in care homes: a systematic
All authors made substantial contributions to conception or design of the review of qualitative studies. Age Ageing. 2012;41(4):429–40.
work, or acquisition of data, analysis or interpretation of data, and were 10. Haugan G. Nurse-patient interaction is a resource for hope, meaning-in-life,
involved in drafting the manuscript or revising it critically for important and self-transcendence in cognitively intact nursding-home patients. Scand
intellectual content. Each author has approved the submitted version (and J Caring Sci. 2014;2014(28):74–8.
any substantially modified version that involves the author’s contribution to 11. Haugan G. Meaning-in-life in nursing-home patients: a valuable approach
the study). All authors read and approved the final manuscript. for enhancing psychological and physical well-being? J Clin Nurs. 2014;
23(13–14):1830–44.
Funding 12. Haugan, G., et al., Development and psychometric properties of the joy-of-life
This study received funding from the Norwegian Council of Research, project scale among cognitively intact nursing home patients. Scand J Caring Sci, 2019.
code 238331. 13. Drageset J, Haugan G, Tranvåg O. Crucial aspects promoting meaning and
purpose in life: perceptions of nursing home residents. BMC Geriatrics,
Open Access. 2017;17(1):254.
Availability of data and materials
14. Duggleby W, et al. Hope, older adults, and chronic illness: a metasynthesis
The datasets generated and/or analysed during the current study are not
of qualitative research. J Adv Nurs. 2012;68(6):1211–23.
publicly available due to Norwegian Act on medical and health research
15. Hammer KOEOC. The meaning of hope in nursing research: a meta-
(ACT 2008–06-20 no. 44):§ 38 but are available from the corresponding
synthesis. Scand J Caring Sci. 2009;23(3):549–57.
author on reasonable request. All raw data is in Norwegian.
16. Reed PG. Theory of self-transcendence, in Middle range theory for nursing.
In: Smith MJ, Liehr PR, editors. . New York: Springer Pub; 2018. p. 119–46.
Ethics approval and consent to participate 17. Haugan G, et al. Self-transcendence in nursing home patients - a resource
Ethical approval was obtained by the Regional Committee for Medical and for well-being. J Adv Nurs. 2013;69(5):1147–60.
Health Research Ethics in Norway (ref.nr 2014/2000/REK Central) as well as by 18. Bakas T, et al. Systematic review of health-related quality of life models.
the Management Units at the 27 nursing homes which participated in this Health Qual Life Outcomes. 2012;10(34).
study. Each participant provided voluntarily written informed consent. 19. Barca ML, et al. Factors associated with depression in Norwegian nursing
homes. Int J Geriatr Psychiatry. 2009;24(4):417–25.
Consent for publication 20. Haugan G. Meaning-in-life in nursing-home patients: a correlate to physical
Each author participated sufficiently in the work to take public responsibility and emotional symptoms. J Clin Nurs. 2014;23(7–8):1030–43.
for appropriate portions of the content. All authors agreed to be personally 21. World Health Organization. Global status report on noncommunicable
accountable for the author’s own contributions and to ensure that questions diseases 2010. 2011. p. 176. ISBN: 978 92 4 156422 9.
related to the accuracy or integrity of any part of the work, even ones in 22. Bowling A, et al. A short measure of quality of life in older age: the
which the author was not personally involved, are appropriately investigated, performance of the brief older People's quality of life questionnaire
resolved, and the resolution documented in the literature. (OPQOL-brief). Arch Gerontol Geriatr. 2013;56(1):181–7.
23. American Psychological Association and National Council on Measurement
Competing interests in Education. Standards for educational and psychological testing.
The authors declare no conflict of interest. Washington, DC: American Educational Research Association; 1999.
24. Goodwin LD, Leech NL. The meaning of validity in the new standards for
Author details educational and psychological testing: implications for measurement
1
NTNU Center for health promotion research, Norwegian University of courses. Meas Eval Couns Dev. 2003;36(3):181–92.
Science and Technology, Trondheim, Norway. 2Faculty of Nursing and Health 25. Netemeyer R, Bearden W, Sharma S. Scaling procedures. Issues and Applications.
Sciences, Nord University, Bodø, Norway. 3Faculty of Health and Social In: Thousand oaks. London, New Dehli: SAGE Publications, Inc.; 2003.
Science, Western University of Applied Science, Bergen, Norway. 4University 26. Ginty, A.T., Psychometric Properties, in Encyclopedia of Behavioral Medicine,
of Bergen, Bergen, Norway. 5Faculty of Nursing, Akdeniz University, Antalya, G. M.D. and T. J.R., Editors. 2013, Springer: New York, NY.
Turkey. 6Social Worker, Kyambogo University and Butabika National Referral 27. Haugan G. The relationship between nurse-patient-interaction and
and Teaching Hospital, Kampala, Uganda. 7Department of Teacher Education, meaning-in-life in cognitively intact nursing-home patients. J Adv Nurs.
Norwegian University of Science and Technology, Trondheim, Norway. 2014;70(1):107–20.
28. Haugan G, Moksnes UK, Espnes GA. Nurse-patient-interaction: a resource for
Received: 6 August 2019 Accepted: 14 November 2019 hope among cognitively intact nursing home patients. J Holist Nurs. 2013;
31(3):152–63.
29. Haugan, G., Self-transcendence, well-being and nurse-patient interaction in
References cognitively inatct nursing home patients, in Faculty of Social Sciences and
1. World Health Organization. World report on ageing and health. 2018 2018, Technology Management, Department of Social Work and Health Science. 2013,
February 5th 2019, July 17th]. Norwegian University of science and technology: Trondheim, Norway p 88.
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 13 of 14

30. Haugan G, Hanssen B, Moksnes UK. Self-transcendence, nurse-patient 54. Reed PG. Demystifying self-transcendence for mental health nursing
interaction and the outcome of multidimensional well-being in cognitively practice and research. Arch Psychiatr Nurs. 2009;23(5):397–400.
intact nursing home patients. Scand J Caring Sci. 2013;27(4):882–93. 55. Reed PG. Toward a nursing theory of self-transcendence: deductive
31. Haugan G, et al. Self-transcendence and nurse-patient interaction in reformulation using developmental theories. Adv Nurs Sci. 1991;13(4):
cognitively intact nursing-home patients. J Clin Nurs. 2012;21:3429–41. 64–77.
32. Haugan G, Moksnes UK, Løhre A. Intra-personal self-transcendence, 56. Haugan G, et al. The self-transcendence scale - an investigation of the
meaning-in-life and nurse-patient interaction: powerful assets for quality of factor structure among nursing home patients. J Holist Nurs. 2012;30(3):
life in cognitively intact nursing home patients. Scand J Caring Sci. 2016; 147–59.
30(4):790–801. 57. Crumbaugh JC. Cross-validation of purpose-in life test based on Frankl's
33. Drageset J, et al. Sense of coherence as a resource in relation to health- concepts. J Individ Psychol. 1968;24:74–81.
related quality of life among mentally intact nursing home residents - a 58. Crumbaugh JC, Henrion R. The PIL test: administration, interpretation, uses
questionnaire study. Health Qual Life Outcomes. 2008;6:85. theory and critique. Int Forum Logother. 1988;11(2):76–88.
34. Drageset J, et al. Emotional loneliness is associated with mortality among 59. Crumbaugh JC, Maholick LT. An experimental study in existentialism: the
mentally intact nursing home residents with and without cancer: a five-year psychometric approach to Frankl’s concept of noogenic neurosis. J Clin
follow-up study. J Clin Nurs. 2013;22(1–2):106–14. Psychol. 1964;20(2):200–7.
35. Drageset J, et al. The impact of social support and sense of coherence on 60. Crumbaugh JC, Maholick LT. Manual of Instructions for The Purpose-in-life test.
health-related quality of life among nursing home residents--a Psychometric Affilitates, in Viktor Frankl Institute of Logotherapy 1969. 1981.
questionnaire survey in Bergen, Norway. Int J Nurs Stud. 2009;46(1):66–76. 61. Crumbaugh JC, Maholick LT. Manual of Instructions for the Purpose-in-Life
36. Haugan G, Innstrand ST, Moksnes UK. The effect of nurse-patient-interaction Test. Illinois: Psychometric Affiliates: (P.O. Box 2852, Saratoga, CA 95070)
on anxiety and depression in cognitively intact nursing home patients. J Murfreesboro: Psychometric Affiliates; 1969.
Clin Nurs. 2013;22(15–16):2192–205. 62. Steger MF. An illustration of issues in factor extraction and indentification of
37. Haugan, G. and S.T. Innstrand, The effect of self-transcendence on dimensionality in psychological assessment data. J Pers Assess. 2006;86:263–72.
depression in cognitively intact nursing home patients. ISRN Psychiatry, 63. Steger MF, et al. The meaning in life questionnaire: assessing the presence
2012. 2012(Article ID 301325): p. 10. of and search for meaning in life. J Couns Psychol. 2006;53:80–93.
38. Drageset J, Espehaug B, Kirkevold M. The impact of depression and sense of 64. Bondevik, M., The Life of the Oldest Old. Studies concerning loneliness,
coherence on emotional and social loneliness among nursing home social contacts, activities of daily living, purpose in life and religiousness, in
residents without cognitive impairment - a questionnaire survey. J Clin Nurs. Department of Public Health and Primary Health Care, Division for Nursing
2012;21(7–8):965–74. Science. 1997, University of Bergen: Bergen.
39. Waltz CF, Strickland OL, Lenz ER. Measurement in nursing and health 65. Bondevik M, Skogstad A. Loneliness, religiousness, and purpose in life in the
research. 3rd ed. New York: Springer; 2005. oldest old. J Relig Gerontol. 2000;11(1):5–21.
40. Clark LA, Watson D. Constructing validity: basic issues in objective scale 66. Haugan G, Moksnes UK. Meaning-in-life in nursing-home patients: a
development. Psychol Assess. 1995;73:309–19. validation study of the purpose-in-life test. J Nurs Meas. 2013;21(2):296–319.
41. Bowling A, Gabriel ZL. Theories of Quality of Life in Older Age. Aging Soc. 67. Flood M, Scharer K. Creativity enhancement: possibilities for succesful aging.
2007;27(827–48). Issues Ment Health Nurs. 2006;27:939–59.
42. Bowling, A. And S. P., Which measure of quality of life performs best in 68. Haugan, G., et al., Development and psychometric properies of the joy-of-life
older age? A comparison of the OPQOL, CASP-19 and WHOQOL-OLD. J scale in cognitively intact nursing home patients. Scand J Caring Sci, 2019.
Epidemiol Community Health, 2010. 65: p. 273–280. 69. Keyes, C.L., Mental health as a complete state: how the salutogenic
43. Bowling, A., Psychometric properties of the older People’s quality of life perspective completes the picture in Bridging Occupational, Organizational
questionnaire validity. Curr Gerontol Geriatr Res, 2009. 298950. and Public Health, G.F. Bauer and O. Hämmig, Editors. 2014, Springer:
44. Bowling A, Stenner P. Which measure of quality of life performs best in Netherlands. p. 179–92.
older age? A comparison of the OPQOL, CASP-19 and WHOQOL-OLD. J 70. Keyes CL. Promoting and protecting mental health as flourishing: a complementary
Epidemiol Community Health. 2011;65:273–80. strategy for improving national mental health. Am Psychol. 2007;62(2):95–108.
45. Bilotta C, et al. Dimensions and correlates of quality of life according to 71. Keyes CL, Lopez SJ. Toward a science of mental health. In: Snyder CR, Lopez
frailty status: a crosssectional study on community-dwelling older adults SJ, editors. Oxford handbook of positive psychology. Oxford: Oxford
referred to an outpatient geriatric service in Italy. Health Qual Life University Press; 2009. p. 89–95.
Outcomes. 2010;8(56). 72. Seligman M. Flourish: a visionary new understanding of happiness and well-
46. Bilotta C, et al. Older People’s Quality of Life (OPQOL) scores and adverse being. New York: Simon and Schuster; 2012.
health outcomes at a one-year follow-up. A prospective cohort study on 73. Smith M. Review of research related to Watson's theory of caring. Nurs Sci
older outpatients living in the community in Italy. Health Qual Life Q. 2004;17(1):13–25.
Outcomes. 2011;9(72). 74. Watson, J., Nursing: human science and human care. A theory of nursing.
47. Bilotta C, et al. Quality of life in older outpatients living alone in the 1988, New York: National League for Nursing.
community in Italy. Health Soc Care Commun. 2012;20:32–41. 75. Nåden D, Sæteren B. Cancer patients’ perception of being or not being
48. Mares J, Cigler H, Vachkova E. Czech version of OPQOL-35 questionnaire: confirmed. Nurs Ethics. 2006;13(3):222–35.
the evaluation of the psychometric properties. Health Qual Life Outcomes. 76. Watson J. Watson's theory of human caring and subjective living
2016;14(93). experiences: carative factors/caritas processes as a disciplinary guide to the
49. Chen Y, Hicks A, While A. Validity and reliability of the modified professional nursing practice. Texto Contexto - Enferm. 2017;16:129–35.
Chinese version of the Older People's Quality of Life Questionnaire 77. Rinnan E, et al. Joy of life in nursing homes: a qualitative study of what
(OPQOL) in older people living alone in China. Int J Older People constitutes the essence of joy of life in elderly individuals living in
Nursing. 2014;9(4):306–16. Norwegian nursing homes. Scand J Caring Sci. 2018.
50. Hendricson W, et al. Development and initial validation of a dual-language 78. Haugan G, Drageset J. The hospital anxiety and depression
English-Spanish format for the arthritis impact measurement scales. Arthritis scale—dimensionality, reliability and construct validity among cognitively
Rheum. 1989;32:1153–9. intact nursing home patients. J Affect Disord. 2014;165:8–15.
51. Beaton, D., et al., Recommendations for the Cross-Cultural Adaptation of the 79. StataCorp. Stata: Release 16. Statistical Software. College Station: StataCorp
DASH and Quick DASH Outcome Measures. 2007, Toronto: : Institute for LLC; 2019.
Work and Health. 80. Hurley AM, et al. Exploratory and confirmatory factor analysis: guidelines,
52. Guillemin F, Bombardier C, Beaton D. Cross-cultural adaptation of health- issues, and alternative. J Organ Behav. 1997;18:667–83.
related quality of life measures: Literature review and proposed guidelines. J 81. Brown T. Confirmatory factor analysis for applied research. New York: The
Clin Epidemiol. 1993;46:1417–32. Guilford Press; 2006.
53. Reed, P.G., Theory of Self-Transcendence, in Middle Range Theory for 82. Sharma S. Applied multivariate techniques. New York: Wiley; 1996.
Nursing, M.J. Smith and P.R. Liehr, Editors. 2008, Springer publishing 83. Tabachnick BG, Fidell LS. Using Multivariate Statistics. 7th ed. Harlow:
company, LLC: New York p 105-129. Pearson Education Inc; 2019.
Haugan et al. Health and Quality of Life Outcomes (2020) 18:1 Page 14 of 14

84. Mehmetoglu, M. and T.G. Jakobsen, Applied Statistics using STATA. A guide
for the social sciences. 2017, Los Angelos - London - New Dehli - Singapore
- Washington DC - Melbourne: SAGE.
85. Mehmetoglu, M. and T. Jakobsen, Applied Statistics using STATA. A
guide for the social sciences. 2017, Los Angelos - London - New
Dehli - Singapore - Washington DC - Melbourne: SAGE.
86. Acock AC. Discovering structural equation modeling using Stata revised ed.
Texas: STATA Press; 2013.
87. Satorra A, Bentler P. Corrections to Test Statistics and Standard Errors in
Covariance Structure Analysis. In: Von Eye A, Cloggs C, editors. Latent
variables analysis: Applications for developmental research. Thousand Oaks:
Sage; 1994. p. 399–419.
88. Kline R. In: Little TD, editor. Principles and Practice of Structural Equation
Modeling. 3rd ed. New York: The Guildford Press; 2011.
89. Osborne, J.W. and A.B. Costello, Best Practices in Exploratory Factor Analysis:
Four Recommendations for Getting the Most From Your Analysis. Practical
Assessment, Research & Evaluation, 2005. 10(7, July 2005): p. 1–9.
90. Hair Jj, et al. Multivariate data analysis. Upper Saddle River: Prentice Hall;
2010.
91. Fabrigar LR, et al. Evaluating the use of exploratory factor analysis in
psychological research. Psychol Methods. 1999;3:272–99.
92. Nunally JC, Bernstein IH. Psychometric theory. New York: McGraw-Hill; 1994.
93. Raykov T. Estimation of congeneric scale reliability using covariance tructure
analysis with nonlinear constraints. Br J Math Stat Psychol. 2001;54(2):315–23.
94. Fayers P, Machin D, editors. Quality of Life. The assessment, analysis and
interpretation of patient-reported outcomes. 2nd ed. Chisester: John Wiley
& Sons Ltd.; 2007.
95. Slettebo A, et al. The significance of meaningful and enjoyable activities for
nursing home resident's experiences of dignity. Scand J Caring Sci. 2017;
31(4):718–26.
96. Grönstedt H, et al. Effects of individually tailored physical and daily activities
in nursing home residents on activities of daily living, physical performance
and physical activity level: a randomized controlled trial. Gerontol. 2013;
59(3):220–9.
97. Brownie S, Horstmanshof L. Creating the conditions for self-fulfilment for
aged care residents. Nurs Ethics. 2012:1–10.
98. Haugland BØ. Meningsfulle aktiviteter på sykehjemmet [Meaningful
activities in nurisng homes]. Sykepleien Forskning. 2012;7(1):42–9.
99. Harper Ice G. Daily life in a nursing home - has it changed in 25 years? J
Aging Stud. 2002;16:345–59.
100. FABBRI E, et al. Aging and Multimorbidity: New Tasks, Priorities, and
Frontiers for Integrated Gerontological and Clinical Research. J Am Med Dir
Assoc. 2015;16(8):640–7.
101. Otsuka S, et al. Prospects for introducing the Eden Alternative to Japan. J
Gerontol Nurs. 2010;36(3):47–55.
102. Choi N, Ransom S, Wyllie R. Depression in older nursing home residents: the
influence of nursing home environmental stressors, coping, and acceptance
of group and individual therapy. Aging Ment Health. 2008;12(5):536–47.
103. Tuckett A. The meaning of nursing-home: ‘waiting to go up to St. Peter, OK!
Waiting house, sad but true’: an Australian perspective. J Aging Stud. 2007;
21(2):119–33.
104. Bergland A, Kirkevold M. Thriving in nursing homes in Norway: contributing
aspects described by residents. Int J Nurs Stud. 2006;43(6):681–91.
105. Bergland A, Kirkevold M. Resident-caregiver relationships and thriving
among nursing home residents. Res Nurs Health. 2005;28(5):365–75.
106. Bollen KA. Structural equations with latent variables. New York: Wiley; 1989.

Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.