Clinical validation of the NANDA-I diagnosis of Impaired Memory in elderly
patients

Michelle H. Montoril MsN, RN, Marcos Venı́cios O. Lopes PhD, RN,

Rosimere F. Santana PhD, RN, Vanessa Emille C. Sousa MS, Priscilla
Magalhães O. Carvalho RN, Camila M. Diniz RN, Naiana P. Alves RN,
Gabriele L. Ferreira RN, Nathaly Bianka M. Fróes RN, Angélica P. Menezes
RN

PII: S0897-1897(15)00168-8
DOI: doi: 10.1016/j.apnr.2015.08.005
Reference: YAPNR 50715

To appear in: Applied Nursing Research

Received date: 9 March 2015

Revised date: 5 August 2015
Accepted date: 14 August 2015

Please cite this article as: Montoril, M.H., Lopes, M.V.O., Santana, R.F., Sousa, V.E.C.,
Carvalho, P.M.O., Diniz, C.M., Alves, N.P., Ferreira, G.L., Fróes, N.B.M. & Menezes,
A.P., Clinical validation of the NANDA-I diagnosis of Impaired Memory in elderly pa-
tients, Applied Nursing Research (2015), doi: 10.1016/j.apnr.2015.08.005

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 ACCEPTED MANUSCRIPT
1

Clinical validation of the NANDA-I diagnosis of Impaired Memory in elderly patients

Michelle H. Montoril, MsN, RNa,

Marcos Venícios O. Lopes, PhD, RNa,
Rosimere F. Santana, PhD RNb
Vanessa Emille C. Sousa, MSa,

PT
Priscilla Magalhães O. Carvalho, RNa,
Camila M. Diniz, RNa,
Naiana P. Alves, RNa,

RI
Gabriele L. Ferreira, RNa,
Nathaly Bianka M. Fróes, RNa,
Angélica P. Menezes, RNa.

SC
a
Nursing Department. Federal University of Ceará. 1115 Alexandre Baraúna st, Rodolfo
Teófilo, Fortaleza, Ceará, Brazil. Postal code: 60430-160.

NU
b
Nursing Department. Fluminense Federal University. R. Dr. Celestino, 74 - Centro, Niterói ,
Rio de Janeiro, Postal Code: 24020-091

Corresponding Author:
MA
Marcos Venícios de Oliveira Lopes, RN, PhD
Federal University of Ceará
Mailing address: 1055, Esperanto St, Vila União. Fortaleza – CE / Brazil. Postal Code:
60410-622. Phone number: 55 85 33668459. Fax number: 55 85 33668456. E-mail:
ED

marcos@ufc.br

Running head: Impaired Memory in elderly patients

PT

Conflict of interest
None.
CE

Funding
This research received no specific grant from any funding agency in the public, commercial,
AC

or not-for-profit sectors.
 ACCEPTED MANUSCRIPT
2

Clinical validation of the NANDA-I diagnosis of Impaired Memory in elderly patients

Abstract

Purpose: To perform a clinical validation of the defining characteristics of Impaired Memory

PT
(IM) in elderly patients at a long-term care institution. Methods: A sample of 123 elderly

RI
patients was evaluated with a questionnaire designed to identify IM according to the

NANDA-I taxonomy. Accuracy measures were calculated for the total sample and for males

SC
and females separately. Results: Sensitivity and specificity values indicated that: (1) Inability

NU
to learn new skills is useful in screening IM, and (2) Forgets to perform a behavior at a

scheduled time, Forgetfulness, Inability to learn new information, Inability to recall events,
MA
and Inability to recall factual information are confirmatory indicators. Conclusion: Specific

factors can affect the manifestation of IM by elderly patients. The results may be useful in
ED

improving diagnostic accuracy and efficiency of the IM nursing diagnosis.

Keywords: Geriatric nursing, Nursing diagnosis, Clinical validation, Memo

PT
CE

Introduction

Aging is a dynamic and progressive process characterized by morphological,

AC

functional, biochemical and psychological changes that can determine the loss of an

individual's ability to adapt to his or her environment (Wold, 2011). One of the largest health

adversities related to aging is cognitive impairment (including impairments in episodic

memory), which causes restriction or loss of skills needed to perform functions and activities

of daily living (Drag and Bieliauskas, 2010). Thus, among numerous specific indicators, the

presence of physical and cognitive deficits determines the health status of the elderly.

The decline of intelligence, learning ability, and memory represents important cognitive

deficits associated with aging (Eliopoulos, 2014). Memory is the most widely studied

cognitive ability, and the decline of this function is a major concern for elderly people
 ACCEPTED MANUSCRIPT
3

(Barcelos-Ferreira and Bottino, 2014). Additionally, this function is essential for the storage

of information, the knowledge of self and the world, the development of language, the

recognition of people, and the consciousness of the continuity of life itself (Volkers and

PT
Scherder, 2014).

Nurses may have a critical eye for individuals who are experiencing specific negative

RI
medical effects during the aging process, and can acquire the necessary skills to carry out an

SC
effective plan of care that contributes to the promotion of quality of life. Impaired Memory is

a nursing diagnosis from the NANDA-I taxonomy II that is embedded in the

NU
Perception/Cognition domain. This diagnosis is defined as the inability to remember or recall
MA
information or behavioral skills, and it includes a set of ten defining characteristics: Forgets

to perform a behavior at a scheduled time, Inability to learn new information, Inability to

ED

learn new skills, Inability to perform a previously learned skill, Inability to recall events,

Inability to recall factual information, Inability to recall if a behavior was performed,

PT

Inability to retain new information, Inability to retain new skills, and Forgetfulness (Herdman

and Kamitsuru, 2014).

CE

There are few studies on the nursing diagnosis of Impaired Memory, and this gap is
AC

even greater in the case of studies focused on the investigation of defining characteristics of

nursing diagnoses. Impaired Memory has been reported in prevalence studies and, in most

cases, is listed together with other nursing diagnoses (Brusamarello, 2013; Güler et al., 2012;

Frauenfelder, 2011). Although exploratory approaches are necessary, they do not allow for

analysis of the relationship between a given nursing diagnosis and its defining characteristics,

which are essential for understanding how the diagnostic process is carried out in specific

situations or populations.

Chaves et al. (2010) developed a content validation of the Impaired Memory

components from the NANDA-I taxonomy II with a panel of experts. However, the authors
 ACCEPTED MANUSCRIPT
4

did not perform a clinical validation of this nursing diagnosis, and suggested it for further

studies. In an exploratory study, Souza and Santana (2011) proposed the clinical validation of

a protocol that was specifically designed for the identification of Impaired Memory, but the

PT
study was intended for hospitalized patients with a diagnosis of dementia and was focused on

the prevalence of the defining characteristics with no investigation of the occurrence of

RI
relationships between the variables.

SC
There are difficulties in establishing clinical criteria that favor the accurate

identification of Impaired Memory. In general, nurses use vague and inconsistent criteria in

NU
clinical practice to identify the presence of cognitive deficits (Carpenito, 2012; Elliott et al.,
MA
2008). Thus, it is not certain that two nurses assessing the same nursing diagnosis are talking

about the same thing, because the diagnostic assessment is made based on personal, instead
ED

of standardized, criteria. As a result, there is confusion in the diagnostic process involving

Impaired Memory and other nursing phenomena (Ried and Dassen, 2000).
PT

Difficulties in the nursing diagnostic process have stimulated the development and

dissemination of accuracy and clinical validation studies, which have become important
CE

research approaches. The accuracy determines the direct relationship between a defining
AC

characteristic and the presence or absence of a given nursing diagnosis. In other words, the

accuracy measurements allow for the establishment of clinical indicators that are appropriate

in the inference of certain diagnoses, which contributes to improved clinical reasoning, the

delimitation of the nursing skills, and consolidated evidence-based clinical practice (Lopes et

al., 2012).

Clinical validation studies are needed not only to increase the validity of the diagnoses

that belong to a particular taxonomy, but also to provide tools to more accurately assess the

health status of patients. The primary aim of this study was to conduct a clinical validation

for the defining characteristics of Impaired Memory in elderly residents of a long-term care
 ACCEPTED MANUSCRIPT
5

institution. As a secondary aim, we examined the presence of relationships between

demographic variables and the accuracy values of the defining characteristics identified in the

sample.

PT
Material and methods

RI
Design and sample

SC
This cross-sectional study was conducted in a long-term care facility for elderly

persons, which specializes in the care and treatment of people with partial or total

NU
dependence, and is situated in northeast Brazil. The sample size was obtained using the
MA
following formula: n = Zα2 * Se * (1-Se)/L2 * P, where “Zα” refers to the 95% confidence

level, “Se” represents an expected sensitivity of 80%, “L” is the half-width of a 95%

confidence interval for conjectured sensitivity of 10%, and “P” is the prevalence for Impaired
ED

Memory of 50%, which is the recommended percentage when the prevalence of an event is
PT

unknown.

A convenience sample of 123 participants was recruited from the long-term care
CE

institution from a total of 222 patients. Participants had to be at least 60 years old to be
AC

included in the study. Exclusion criteria included the presence of verbal or cognitive deficits,

and / or cognitive bewilderment that could prevent interaction with the subjects or lead to

data collection errors. To find subjects who fit the profile, a nurse working at the institution

was consulted during the recruitment period and asked about potential subjects’ participation

in the study. This nurse was also consulted after data collection to validate the subjects'

responses. We used this recruitment strategy because this nurse had worked at the institution

for years and knew each subject well.

 ACCEPTED MANUSCRIPT
6

Written consent was obtained from each participant after the purpose, nature, and

potential complications of the study were explained. No incentives were given to the subjects.

Ethical approval was obtained from the institutional review board.

PT
Data collection

The data collection form included demographic data (gender, age, family income,

RI
education, and length of stay) and questions related to the defining characteristics of Impaired

SC
Memory according to the NANDA-I taxonomy II (Herdman and Kamitsuru, 2014). Previous

studies have found cognitive deficit, including memory impairment, among elderly with an

NU
income of ⩽$5000 and projections suggest an increasing proportion of cognitive problems
MA
and dementia in less developed countries (National Institute on Aging, 2011; Berkman et al.,

1993). Furthermore, cross-sectional and longitudinal results showed that, in less educated
ED

elderly, memory decline is faster and sets in at an earlier age (Schmand et al., 1997; Berkman

et al., 1993). In addition, another study found differences in lengths of stay were statistically
PT

significant after adjusting for age, race, and gender for elderly with cognitive impairment

(Lyketsos et al., 2000).

CE

Five trained nurses collected the data independently and individually through guided
AC

interviews. Inter-rater reliability was not measured; however, the principal investigator

supervised the nurses’ data collection with the first 20 subjects to ensure that they were using

the same procedures. It is worth mentioning that the nurses were responsible only for the

memory assessment; the presence of the nursing diagnosis Impaired Memory was estimated

by the latent class analysis (LCA) method, which is described further (see the data analysis

section).

All nurses were trained and instructed to use a standard operating procedure form with

operational definitions and a description of how to assess each defining characteristic. The

training lasted three hours and included the definition of Impaired Memory and its
 ACCEPTED MANUSCRIPT
7

components, the physiological and psychosocial aspects of this nursing diagnosis, and the

context of the study population. The operational definitions were adapted from a study that

validated a protocol for the identification of Impaired Memory in hospitalized elderly (Souza

PT
and Santana, 2011).

Specific standard operating procedures were developed to guide the assessment of each

RI
defining characteristic. The memory assessment was composed mostly by questions directly

SC
asked to the patients, such as whether they performs daily activities at a scheduled time and if

they remember the full name of their children or parents. All the patient's answers were

NU
validated by a nurse, who was also asked about some questions (e.g. if the patient has
MA
forgotten specific events or previously performed actions). The assessment also involved the

evaluation of the ability to learn and retain skills/information by teaching the patient how to
ED

unlock the screen of a mobile phone, teaching the name of unknown objects, and applying the

clock draw test.

PT

A pre-test was performed with five subjects who were recruited using the same

selection criterion and no changes were made in the collection form after that. This pre-test
CE

was also used to verify whether the team members were able to perform a standardized data
AC

collection.

Data analysis

Data analysis was performed using the R software package version 3.0.2 (R Core

Team, 2014). Descriptive statistics including percentage, mean, and standard deviations were

obtained. The Jarque-Bera test was applied to verify the normality of the data. The Mann-

Whitney test was applied to verify differences of age / schooling level and the presence /

absence of defining characteristics. The association between gender and defining

characteristics was evaluated using the Chi squared test with Yates’ correction. The Yates

correction is an adjustment made to account for the fact that Pearson’s chi-square test is
 ACCEPTED MANUSCRIPT
8

biased upwards for a 2 x 2 contingency table. This correction is usually recommended if the

expected cell frequencies are below 5.

The accuracy analysis was based on sensitivity and specificity measures of each

PT
defining characteristic of Impaired Memory. The LCA method was applied to calculate the

accuracy measures and to estimate the presence of Impaired Memory in the sample. This

RI
method is employed when no perfect reference standard is present (Qu et al., 1996), and it is

SC
based on the assumption that an unobserved or latent variable (nursing diagnosis) determines

the associations between observable variables (defining characteristics). A two latent class

NU
model with random effects was fitted to obtain estimates of sensitivity, specificity, and their
MA
respective 95% confidence intervals (95% CI). The random effects model assumes a

conditional dependence between the defining characteristics, included in the latent class

analysis. The likelihood ratio chi-square (G2) statistics were applied to help determine how
ED

well the particular models fit the data.

PT

An initial latent class model including all ten defining characteristics was initially set

and called the “null model”. From this model, statistically insignificant defining
CE

characteristics were identified and excluded, and a new model was adjusted. Absence of
AC

statistical significance was defined when the upper limit of the sensitivity and specificity

confidence intervals were less than 50% and/or when the confidence interval included this

value. For all other analyses, we adopted the 5% level of significance (p-value < 0.05).

Results

The study participants were 68 male (55.28%) and 55 female (44.72%) older adults

with an average age of 74.21 years (SD = 7.73) and a median of 2 years of education. Half of

the sample had been living in the institution for over 3 years (Table 1). The most prevalent

clinical indicators of Impaired Memory were: Inability to recall if a behavior was performed
 ACCEPTED MANUSCRIPT
9

(67.21%), Inability to retain new information (59.35%), and Inability to perform a previously

learned skill (55.28%). The estimated prevalence of Impaired Memory in the sample was

31.05%.

PT
After examining the relationships between demographic variables and the accuracy

values of the defining characteristics, significant differences by gender were found. We

RI
decided to conduct further analysis of this relationship using bivariate analysis, which is

SC
shown in Table 2. A statistically significant result (p <0.05) was found between gender and

the defining characteristics Inability to learn new skills, Inability to perform a previously

NU
learned skill, and Inability to retain new skills, with females demonstrating a more
MA
unfavorable outcome than males. Elderly females were approximately three times more likely

to present these defining characteristics than males.

ED

Table 3 shows the bivariate analysis of the relationship between the defining

characteristics of Impaired Memory and the variables age and education level. The incidence
PT

of the defining characteristics Forgetfulness, Inability to learn new skills, and Inability to

retain new skills was higher in the older participants (80 vs. 74; 78 vs. 72; and 79 vs. 73,
CE

respectively). In addition, low schooling level was associated with a high incidence of the
AC

following defining characteristics: Forgetfulness (0 vs. 3; p = 0.019), Inability to perform a

previously learned skill (1 vs. 5; p < 0.001), Inability to recall events (1 vs. 3; p = 0.041), and

Inability to recall factual information (1 vs. 3; p = 0.006).

Table 4 shows the results for the sensitivity and specificity of the defining

characteristics of Impaired Memory from three fitted latent class models. According to these

models, the estimated prevalence of Impaired Memory was between 31% and 36.4%. In the

latent class model that included the whole sample, statistical significance was found for all

defining characteristics, except for Inability to recall if a behavior was performed. The
 ACCEPTED MANUSCRIPT
10

majority of the defining characteristics had specificity values above 90%, and low sensitivity

values.

In the latent class model that included only males, no statistical significance was found

PT
regarding the following defining characteristics: Inability to recall if a behavior was

performed, and Inability to retain new information. Inability to perform a previously learned

RI
skill was the only characteristic with high specificity and sensitivity values in males. The

SC
remaining defining characteristics had only high specificity values.

In the latent class model that included only females, no statistical significance was

NU
found regarding one defining characteristic: Inability to recall if a behavior was performed.
MA
Five defining characteristics had only high specificity values, and two had only high

sensitivity values. Additionally, the defining characteristics Inability to learn new skills and
ED

Inability to retain new skills had high values for both sensitivity and specificity.
PT

Discussion

Clinical validation studies are important for the refinement of diagnostic taxonomies
CE

such as NANDA-I (Sousa et al., 2015). The contribution of this type of study is even more
AC

significant when it involves poorly researched diagnoses or specific populations such as

Impaired Memory in the elderly.

The estimated prevalence of Impaired Memory in the sample was 31.05%, which is

slightly less than 39.4% (Souza and Santana, 2011) and 45.9% (Güler et al., 2012), which are

reported in other studies about this nursing diagnosis. As previously mentioned, the existence

of cognitive impairments/bewilderment was added as an exclusion criterion in this study to

avoid data collection errors. Although this was a necessary measure, it may have contributed

to the low incidence rate of Impaired Memory in the sample.

 ACCEPTED MANUSCRIPT
11

The most prevalent defining characteristics were Inability to recall if a behavior was

performed, Inability to retain new information and Inability to perform a previously learned

skill. The percentages for the presence of these defining characteristics in our study were very

PT
similar to the findings from Souza and Santana (2011) (67.21% and 62.5%, 59.35% and

66.7%, and 55.28% and 58.3%, respectively). In addition to this, Inability to recall if a

RI
behavior was performed and Inability to retain new information were considered defining

SC
characteristics of “greater importance” in a content validation study for the elements of

Impaired Memory (Chaves et al., 2010).

NU
The literature suggests that mild cognitive decline is a consequence of the physiological
MA
processes of aging (Petersen, 2004; Burns and Zaudig, 2002). Thus, the occurrence of the

mentioned defining characteristics in the sample appears to reflect the physiological process
ED

that the participants are experiencing.

The defining characteristic Inability to retain new information is related to short-term

PT

memory, which declines naturally with age, even when intellectual capabilities are preserved.

Inability to recall if a behavior was performed and Inability to perform a previously learned
CE

skill are related to difficulty in recovering stored information, which is also associated with
AC

the aging process (Barcelos-Ferreira and Bottino, 2014; Elliott et al., 2008).

In the analysis of associations between variables, it was found that some defining

characteristics of Impaired Memory are more likely to occur in women, the oldest

individuals, and those with a poor education level. Although some studies have shown that

gender differences may affect perception and attention, conclusions about the relationship

between gender and cognitive abilities are only speculative (Halpern, 2011). In a previous

study, authors discussed gender differences using several parameters, including cognition,

and concluded that the performance of men and women was similar (Wolbers and Hegarty,

2010).
 ACCEPTED MANUSCRIPT
12

Age is indicated as an intrinsic risk factor for cognitive decline (Almeida et al., 2013),

which explains why Forgetfulness was more prevalent in relatively older people in the study

sample. In addition to that, older people typically have less readiness to learn. Thus, elderly

PT
are more dependent on previous experiences than trying new techniques when solving

problems (Eliopoulos, 2014). In addition, with aging, differences in the intensity and duration

RI
of physiological arousal of the individual are common, leading to a loss of remembering

SC
previous answers and of acquisition of new information (Wold, 2011). This may be related to

the occurrence of the defining characteristics related with learning in the sample.

NU
Regarding education, the study sample had a median of 2 years of formal education.
MA
Although such a low educational level is not typical for the geographical area (the fourth

largest metropolitan area in Brazil), it is typical for the group included in the sample, as the
ED

study was carried out in a philanthropic institution. Most of the elders living in this institution

came from the Northeast countryside of Brazil and have not only received a poor education,
PT

but have lived most of their lives with very restricted resources.

The educational level of the patients could have influenced the results, for example, by
CE

contributing to the presence of some clinical indicators that are not exclusive to Impaired
AC

Memory but that are also related to the educational level (for example, Inability to learn new

skills/information). At the same time, this does not invalidate the findings because it is

known that good cognitive performance (including memory) depends on stimuli received

throughout life (e.g., at school) (Wolbers and Hegarty, 2010).

The relationship between educational level and cognitive performance is noted in

several studies. Brucki and Nitrini (2014) reported that individuals with low educational level

are exposed to risk factors for Dementia, such as low cognitive reserve, poor control of

cerebrovascular risk factors, difficulties during cognitive evaluation, and poor cognitive test

adaptations. Low educational level is also presented as one of the frailty predictors found in a
 ACCEPTED MANUSCRIPT
13

study about the impact of social, demographic, health, cognitive, functional, and psychosocial

variables on the cognitive status of community-dwelling elderly (Neri et al., 2013). Authors

of a recent systematic review showed that the dementia risk increases with decreasing

PT
educational level: dementia risk was 2.61 higher among subjects with low schooling

compared to subjects with higher educational level (Meng and D'Arcy, 2012).

RI
Learning influences the functional organization of the human brain. Low education

SC
changes the pattern of brain potentials related to remembering events (Angel et al., 2010), as

well as the hippocampus activation of the right posterior insula, thalamus and operculum

NU
when performing memory tasks (Stern et al., 2005).
MA
It should be noted that lack of motivation is also a factor that may have contributed to

the findings related to Inability to learn new skills and Inability to retain new skills. Although
ED

this factor was not investigated in the study, it was observed that many seniors remained in

their accommodations for long periods, without interacting with family members or third
PT

parties, which may reflect a situation of low stimulation for the development of cognitive

skills.
CE

The accuracy analysis by sensitivity and specificity measures indicated that some
AC

defining characteristics had high levels of both sensitivity and specificity, such as Inability to

perform a previously learned skill (male), Inability to learn new skills (total sample), and

Inability to retain new skills (female). Such defining characteristics are good clinical

indicators of Impaired Memory, and may be useful for diagnostic screening.

However, the defining characteristics that had high specificity values, such as Forgets

to perform a behavior at a scheduled time, Forgetfulness, Inability to learn new information,

Inability to recall events and Inability to recall factual information, are important as

confirmatory indicators of Impaired Memory after the diagnostic screening because high

specificity relates to the ability to correctly exclude a clinical condition (Lopes et al.,2012).
 ACCEPTED MANUSCRIPT
14

It is important to notice that the discriminatory value of a diagnostic test for screening

is reduced when the condition of interest has a low prevalence, even with high sensitivity and

specificity. Thus, it is wise to correlate the accuracy measures of the defining characteristics

PT
with its incidence rates in the study. Considering this correlation, Inability to perform a

previously learned skill can be seen as the most accurate indicator of Impaired Memory in

RI
this study.

SC
The significance of this study is its contribution to the knowledge needed for better

assessment of memory impairment in elderly patients. As previously stated, nursing

NU
diagnoses from the Perception/Cognition Domain can be difficult to assess, and this difficulty
MA
can lead to either a misevaluation of the patients' cognitive functions or neglect of the human

responses that belongs to this Domain. It has been recognized that every mental process
ED

involves some aspect of memory; hence, more emphasis should be given to memory

assessment in elderly patients.

PT

Although the aim of this study is not to review the Impaired Memory components

proposed by the NANDA-I, our results indicate that the refinement and improvement of some
CE

components of this diagnosis are needed, especially in relation to some defining

AC

characteristics that are not specifically related to memory, such as Inability to learn new skills

or information.

Finally, it is important to emphasize that the study results should be viewed with

discretion because the participants had relatively low education, old age, and were living in a

long-term care institution for more than three years. Thus, the results may differ among

elderly patients living at home or with a higher educational level.

Conclusions
 ACCEPTED MANUSCRIPT
15

This study enabled the clinical validation of all defining characteristics of Impaired

Memory included in the NANDA-I taxonomy, except for Inability to recall if a behavior was

performed, which had no clinical value in the sensitivity and specificity analysis.

PT
Additionally, it was found that some defining characteristics were not useful for identifying

or excluding Impaired Memory in certain subgroups, such as Inability to retain new

RI
information for males and Inability to recall if a behavior was performed for females.

SC
The defining characteristics that have both high sensitivity and high specificity can

be considered good clinical indicators for Impaired Memory screening. These are: Inability to

NU
perform a previously learned skill (male), Inability to learn new skills (total sample), and
MA
Inability to retain new skills (female). In turn, defining characteristics that have only high

specificity, such as Forgets to perform a behavior at a scheduled time, Forgetfulness, Inability

ED

to learn new information, Inability to recall events, and Inability to recall factual information,

are important as confirmatory indicators of Impaired Memory after the diagnostic screening.
PT

The identification of good clinical indicators, that is, signs and symptoms that can be

used for diagnostic screening and confirmation, contribute to a decrease in errors by nurses,
CE

and enable the evaluation of patients in a more standardized way. Thus, the study findings
AC

may contribute to an early and accurate identification of Impaired Memory in elderly, and

present value in clinical practice.

Finally, the findings should be used with caution, given that these are related to

specific characteristics of the sample such as age, level of education, and psychosocial and

cultural factors. Further studies with elderly being assisted in other settings are needed to

enable a comparison of the predictive ability of the Impaired Memory defining characteristics

in different contexts.
 ACCEPTED MANUSCRIPT
16

References

Almeida, R., Marques, M., Santo, H.E., Moitinho, S., Vigário, V., Pena, I. et al. Selective

attention and cognitive decline in institutionalized elderly. European Psychiatry. 2013;

PT
28: 1. DOI: http://dx.doi.org/10.1016/S0924-9338(13)76308-9.

Angel, L., Fay, S., Bouazzaoui, B., Baudouin, A., and Isingrini, M. Protective role of

RI
educational level on episodic memory aging: an event-related potential study. Brain and

SC
Cognition. 2010; 74: 312-323.

Barcelos-Ferreira, R., and Bottino, C. Prevalence of amnestic mild cognitive impairment

NU
in depressed and nondepressed elderly Brazilian community residents. Alzheimer’s &
MA
Dementia. 2014; 10: 907. DOI: http://dx.doi.org/10.13140/2.1.2420.6724

Berkman, L.F, Seeman, T.E., Albert, M., Blazer, D., Kahn, R., Mohs, R. et al. High, usual and

impaired functioning in community-dwelling older men and women: Findings from the
ED

MacArthur Foundation Research Network on successful aging. Journal of Clinical Epidemiology.

PT

1993; 46: 1129-1140. DOI: http://dx.doi.org/10.1016/0895-4356(93)90112-E.

Brucki, S.M.D., and Nitrini, R. Cognitive impairment in individuals with low educational
CE

level and homogeneous sociocultural background. Dementia & Neuropsychologia. 2014;

8: 345-350.
AC

Brusamarello, T., Capistrano, F.C., Oliveira, V.C., Mercês, N.N.A., and Maftum, M.A.

Caring people with mental disorders and their family members: diagnoses and

interventions from nursing consultation. Cogitare Enfermagem. 2013; 18: 245–252. DOI:

http://dx.doi.org/10.5380/ce.v18i2.32574.

Burns, A., and Zaudig, M. Mild cognitive impairment in older people. Lancet. 2002; 360:

1963–1965.

Carpenito, L. Nursing diagnosis: application to clinical practice. 14th ed. Lippincott,

Philadelphia, PA; 2012.

 ACCEPTED MANUSCRIPT
17

Chaves, E.H.B., Barros, A.L.B.L., and Marini, M. Aging as a related factor of the nursing

diagnosis Impaired Memory: content validation. International Journal of Nursing

Terminologies and Classifications. 2010; 21: 14–20. DOI: http://dx.doi.org/10.1111/j.1744-

PT
618X.2009.01140.x.

Drag, L.L., and Bieliauskas, L.A. Contemporary review 2009: cognitive aging. Journal of

RI
Geriatric Psychiatry and Neurology. 2010; 23: 75–93. DOI:

SC
http://dx.doi.org/10.1038/nrn2298

Eliopoulos, C. Gerontological Nursing. 8th ed. Lippincott Williams & Wilkins,

Philadelphia, PA; 2014.

NU
MA
Elliott, A.F., Horgas, A.L., and Marsiske, M. Nurses' role in identifying mild cognitive

impairment in older adults. Geriatric Nursing. 2008; 29: 38–47. DOI:

ED

http://dx.doi.org/10.1016/j.gerinurse.2007.04.015.

Frauenfelder, F., Müller‐Staub, M., Needham, I., and Achterberg, T.V. Nursing phenomena
PT

in inpatient psychiatry. Journal of Psychiatric and Mental Health Nursing. 2011; 18: 221–

235. DOI: http://dx.doi.org/10.1111/j.1365-2850.2010.01659.x.

CE

Güler, E.K., Eşer, İ., Khorshid, L., and Yücel, Ş.Ç. Nursing diagnoses in elderly residents
AC

of a nursing home: a case in Turkey. Nursing Outlook. 2012; 60: 21–28. DOI:

http://dx.doi.org/10.1016/j.outlook.2011.03.007.

Halpern, D.F. Sex differences in cognitive abilities. Psychology press, Oxford, UK; 2011.

Herdman, T.H., and Kamitsuru, S. NANDA International nursing diagnoses: definitions

& classification 2015-2017. Wiley-Blackwell, Oxford, UK; 2014.

Lopes, M.V.O., Silva, V.M., and Araujo, T.L. Methods for establishing the accuracy of

clinical indicators in predicting nursing diagnoses. International Journal of Nursing

Knowledge. 2012; 23: 134–139. DOI: http://dx.doi.org/10.1111/j.2047-3095.2012.01213.x

 ACCEPTED MANUSCRIPT
18

Lyketsos, C.G., Sheppard, J.M.E., and Rabins, P.V. Dementia in Elderly Persons in a

General Hospital. The American Journal of Psychiatry. 2000; 157: 704-707.

Meng, X., and D'Arcy, C. Education and dementia in the context of the cognitive reserve

PT
hypothesis: a systematic review with meta-analyses and qualitative analyses. PLoS One.

2012; 7: e38268. DOI: http://dx.doi.org/10.1371/journal.pone.0038268.

RI
National Institute on Aging. National Institutes of Health. Global Health and Aging. NIH

SC
Publication no. 11-7737. 2011.

Neri, A.L., Yassudall, M.S., Araújo, L.F., Eulálio, M.C., Cabral, B.E., Siqueira, M.E.C. et al.

NU
Methodology and social, demographic, cognitive, and frailty profiles of community-
MA
dwelling elderly from seven Brazilian cities: the FIBRA Study. Cadernos de Saúde

Pública. 2013; 29: 778–792.

ED

Petersen, R.C. Mild cognitive impairment as a diagnostic entity. Journal of Internal

Medicine. 2004; 256: 183–194.

PT

Qu, Y., Tan, M., and Kutner, M.H. Random effects models in latent class analysis for

evaluating accuracy of diagnostic tests. Biometrics. 1996; 52: 797–810.

CE

R Core Team. R: A Language and Environment for Statistical Computing. (Retrieved

AC

November 12, 2014, from) http://www.r-project.org/; 2014.

Ried, S., and Dassen, T. Chronic confusion, dementia, and impaired environmental

interpretation syndrome: a concept comparison. International Journal of Nursing

Terminologies and Classifications. 2000; 11: 49–59. DOI: http://dx.doi.org/10.1111/j.1744-

618X.2000.tb00390.x

Schmand, B., Smit, J., Lindeboom, J., Smits, C., Hooijer, C., Jonker, C. et al. Low education is a

genuine risk factor for accelerated memory decline and dementia. Journal of Clinical

Epidemiology. 1997; 50: 1025-1033. DOI: http://dx.doi.org/10.1016/S0895-4356(97)00121-2.

 ACCEPTED MANUSCRIPT
19

Sousa, V.E.C., Lopes, M.V.O., and Silva, V.M. Systematic review and meta-analysis of the

accuracy of clinical indicators for ineffective airway clearance. Journal of Advanced

Nursing. 2015; 71: 498-513. DOI: http://dx.doi.org/10.1111/jan.12518

PT
Souza, P.A., and Santana, R.F. The nursing diagnosis, Impaired Memory, in hospitalized

elderly. Acta Paulista de Enfermagem. 2011; 24: 36–42. DOI:

RI
http://dx.doi.org/10.1590/S0103-21002011000100005.

SC
Stern, Y., Habeck, C., Moeller, J., Scarmeas, N., Anderson, K.E., Hilton, H.J. et al. Brain

networks associated with cognitive reserve in healthy young and old adults. Cerebral

Cortex. 2005; 15: 394-402.

NU
MA
Volkers, K.M., and Scherder, E.J.A. Physical performance is associated with working

memory in older people with mild to severe cognitive impairment. BioMed Research
ED

International. 2014; Article ID 762986: 10 pages. DOI:

http://dx.doi.org/10.1155/2014/762986
PT

Wolbers, T., and Hegarty, M. What determines our navigational abilities? Trends in

Cognitive Sciences. 2010; 14: 138–146. DOI: http://dx.doi.org/10.1016/j.tics.2010.01.001.

CE

Wold, G. Basic geriatric nursing. 5th ed. Mosby, St. Louis, MI; 2011.
AC
 ACCEPTED MANUSCRIPT
20

Table 1 – Demographic and socioeconomic characteristics of the sample.

Variables n % 95%CI
Gender
Male 68 55.28 46.06 – 64.25
Female 55 44.72 35.75 – 53.94
Mean (SD) Median (IQR) p value a

PT
Age 74.21 (7.73) 74.00 (13) 0.175
Length of stay (years) 5.84 (7.80) 3.25 (6) <0.001
Income 812.3 (677.0) 724.0 (0) <0.001

RI
Years of education 3.93 (4.52) 2.00 (5) <0.001
a
Jarque-Bera test for normality.

SC
NU
MA
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT
21

Table 2 – Bivariate analysis for defining characteristics of Impaired Memory by gender.

Gender
Defining Characteristics Statistics
Female Male
1. Forgets to perform a behavior at a scheduled time
Presence 1 0 p = 0.447a
Absence 54 68 OR = --

PT
2. Forgetfulness
Presence 4 6 p = 1.000a
Absence 51 62 OR =0,81 (0.16-3.64)

RI
3. Inability to learn new skills
Presence 24 13 p = 0.005b

SC
Absence 31 55 OR = 3.24(1.36-8.01)
4. Inability to learn new information
Presence 1 3 p = 0.627a
Absence 54 65 OR = 0.40(0.01-5.20)

NU
5. Inability to recall if a behavior was performed
Presence 34 48 p = 0.338b
Absence 21 19 OR =0.64(0.28-1.47)
MA
6. Inability to perform a previously learned skill
Presence 40 28 p < 0.001b
Absence 15 40 OR = 3.77 (1.67-8.85)
7. Inability to recall events
p = 0.212a
ED

Presence 7 4
Absence 47 64 OR = 2.37 (0.56-11.68)
8. Inability to recall factual information
Presence 11 10 p = 0.593b
PT

Absence 44 58 OR = 1.44(0.51-4.18)
9. Inability to retain new skills
Presence 21 12 p = 0.019b
CE

Absence 34 56 OR = 2.86(1.17-7.25)
10. Inability to retain new information
Presence 36 37 p = 0.291b
AC

Absence 19 31 OR = 1.58(0.72- 3.54)

a
Fisher's exact test. b Chi-squared test with Yates' correction. OR – Odds ratio.
 ACCEPTED MANUSCRIPT
22

Table 3 – Mann-Whitney tests for differences in age / education level between elderly with
the presence or absence of defining characteristics of Impaired Memory.
Defining characteristics Age Schooling
level
1. Forgets to perform a behavior at a scheduled time 0.199 0.213
2. Forgetfulness 0.033 0.019

PT
3. Inability to learn new skills <0.001 0.182
4. Inability to learn new information 0.898 0.265
5. Inability to recall if a behavior was performed 0.682 0.468

RI
6. Inability to perform a previously learned skill 0.142 <0.001
7. Inability to recall events 0.146 0.041

SC
8. Inability to recall factual information 0.254 0.006
9. Inability to retain new skills 0.002 0.541
10. Inability to retain new information 0.212 0.108

NU
MA
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT
23

Table 4 – Sensitivity (Se) and Specificity (Sp) of defining characteristics for Impaired
Memory based on three latent class models.
Model #1: total sample Se (95%CI) Sp(95%CI)
1. Forgets to perform a behavior at a scheduled 02.62 (00.00 - 100.00 (98.95 -
time 08.68) 100.00)
2. Forgetfulness 17.94 (05.81 - 96.29 (91.41 -

PT
32.89) 100.00)
3. Inability to learn new skills 85.94 (71.39 - 95.07 (87.62 -
100.00) 100.00)

RI
4. Inability to learn new information 10.47 (01.95 - 100.00 (98.84 -
22.13) 100.00)

SC
6. Inability to perform a previously learned skill 83.62 (68.61 - 57.47 (46.53 -
96.30) 68.42)
7. Inability to recall events 22.35 (09.71 - 97.06 (92.66 -
38.19) 100.00)

NU
8. Inability to recall factual information 39.93 (25.15 - 93.22 (86.65 -
56.76) 98.39)
9. Inability to retain new skills 80.33 (63.75 - 97.26 (91.34 -
MA
100.00) 100.00)
10. Inability to retain new information 81.90 (68.08 - 50.80 (39.58 -
94.57) 62.23)
Impaired Memory estimated prevalence: 31.05% G2 = 71.53; DF = 105; p = 0.995
ED

Model #2: male Se (95%CI) Sp(95%CI)

1. Forgets to perform a behavior at a scheduled 00.00 (00.00 - 100.00 (100.00 -
time 00.00) 100.00)
PT

25.61 (07.43 - 100.00 (94.45 -

2. Forgetfulness
52.12) 100.00)
55.49 (34.77 - 100.00 (92.25 -
3. Inability to learn new skills
CE

100.00) 100.00)
12.81 (00.00 - 100.00 (96.36 -
4. Inability to learn new information
33.31) 100.00)
72.24 (50.98 - 75.15 (59.30 -
AC

5. Inability to perform a previously learned skill

100.00) 88.47)
17.07 (00.00 - 100.00 (95.92 -
6. Inability to recall events
37.43) 100.00)
33.42 (14.09 - 95.13 (85.67 -
7. Inability to recall factual information
63.01) 100.00)
49.97 (30.60 - 99.34 (91.62 -
8. Inability to retain new skills
81.49) 100.00)
2
Impaired Memory estimated prevalence: 34.45% G = 33.77; DF = 50; p = 0.962
Model #3: female Se (95%CI) Sp(95%CI)
1. Forgets to perform a behavior at a scheduled 04.99 (00.00 - 100.00 (100.00 -
time 16.84) 100.00)
2. Forgetfulness 19.96 (04.74 - 100.00 (100.00 -
39.21) 100.00)
3. Inability to learn new skills 100.00 (90.67 - 88.67 (74.92 -
100.00) 100.00)
4. Inability to learn new information 04.99 (00.00 - 100.00 (100.00 -
18.07) 100.00)
 ACCEPTED MANUSCRIPT
24

5. Inability to perform a previously learned skill 87.65 (70.64 - 35.83 (20.59 -

100.00) 51.98)
6. Inability to recall events 29.94 (10.63 - 97.05 (90.22 -
52.33) 100.00)
7. Inability to recall factual information 44.90 (22.33 - 94.27 (85.31 -
69.86) 100.00)

PT
8. Inability to retain new skills 91.27 (72.84 - 92.25 (81.24 -
100.00) 100.00)
9. Inability to retain new information 86.76 (68.18 - 46.76 (30.02 -

RI
100.00) 63.22)
2
Impaired Memory estimated prevalence: 36.43% G = 38.09; DF = 37; p =0.419
G2 – Likelihood ratio statistic. DF – Degrees of freedom.

SC
NU
MA
ED
PT
CE
AC
 ACCEPTED MANUSCRIPT
25

Highlights

- The study examines clinical indicators in elderly patients to clinically validate the nursing diagnosis

of Impaired Memory.

PT
- Some differences in occurrence clinical indicators were observed for gender, age, and schooling

level.

RI
- Six indicators that exhibited best measures of diagnostic accuracy based on latent class analysis with

SC
random effects.

NU
- Some differences in measures of diagnostic accuracy were observed for gender.
MA
ED
PT
CE
AC