Nursing Research: Mental Status Measurement: The Mini-Mental State Examination

Individualized assessment of cognitive status is necessary for the planning and evaluation of nursing care to determine the patient’s capacity to understand instructions, be an active participant in his/her care, make health care decisions, and detect changes that will determine subsequent nursing actions. It is especially important to assess the cognitive status of elders who may have an undetected mild cognitive impairment or delirium; for example, assessing baseline cognitive status of hospitalized elders would allow early detection of side effects from receiving a new medication or of
postoperative delirium. The Mini-Mental State Examination (MMSE) (Folstein, Folstein, & McHugh, 1975) is frequently used as a clinical assessment in a variety of settings and for preliminary screening of elders for neurodegenerative disorders such as Alzheimer’s disease (AD). The MMSE was developed over 30 years ago for serial testing of the cognitive mental state of patients on a neuro geriatric ward.

 Mental Status Measurement

The MMSE was specifically developed to be a formal and relatively thorough clinical evaluation tool that is brief and easy to administer, and consists of eleven tasks of cognition:
(1) orientation to time and place, (2) registration assessed by ability to learn the names of three unrelated objects, (3) attention and calculation by performing serial sevens or spelling the word “world” backwards, (4) recall by naming the three objects previously learned, (5) language assessed by six items of naming objects, following verbal and written commands, writing a sentence spontaneously, and (6) visual-spatial ability by copying two intersecting pentagons. The tester asks the patient to respond to items and records the score for each. Item scores are added to provide the MMSE score, which can range from 30 (all correct) to 0 (no correct) responses.
Before conducting an MMSE assessment, the nurse should make the patient comfortable and establish rapport. The test is not timed but usually takes 5–10 minutes. During the testing, the nurse should praise success and attempt to enhance cooperation by not pressing on items the patient finds difficult. The testing situation may be an embarrassment for patients who are aware that they are “missing” some of the items (Mahoney, Volicer, &Hurley, 2000) and the nurse needs to be sensitive to this phenomenon and protect the self-esteem of such patients while preserving the integrity of the testing procedures to assure administration accuracy.
The degree to which the MMSE is reliable and valid is critical to the interpretation of scores. The tester needs to follow the administration procedures exactly. The MMSE is considered to have satisfactory psychometric properties. Initial reliability and accuracy by measuring consistency in the items and different raters were adequate for interrater agreement and retest stability when two samples of patients and several test administrators were compared (Folstein, Folstein, & McHugh, 1975). Validity, the degree to which the MMSE measures the construct of cognitive impairment, was supported by convergent and discriminant validity comparing hypothesized similarities/differences between scores from three groups of participants: normal, demented, and depressed subjects with and without cognitive symptoms. Age and education may influence test scores (Butler,S. M., Ashford, & Snowdon, 1996) in that elders and persons with low education may score slightly lower yet have higher cognitive capacity, so there are MMSE test norm based on these variables (Crum, Anthony, Bassett, & Folstein, 1993). After many yearof use in several studies, a score of 23 points or less is considered to be preliminary evidence of cognitive impairment and grounds for further evaluation (Cockrell & Folstein,1988).
Clinically, scores on the MMSE should be considered with other assessment data to provide clinicians with an estimate of patients’ cognitive capacity to make treatment and research decisions. MMSE scores are one of several neuropsychological test results used to assess the need for and/or efficacy of cognitive enhancing medications for persons with AD. In the research arena, the MMSE is useboth as an enrollment criterion as well as to describe project participants, enabling comparisons across studies.
At least reliability and if possible validity estimates of instruments used in particular studies should be reported each time the instrument is used in a study. Over 30 years use of the MMSE in reported studies illustrates the stability of the initially reported psychometrics. The MMSE has been translated into several languages and modifications have been made for versions that are culturally and linguistically appropriate (Folstein, 1998).Our research team has used the MMSE in all our studies conducted since 1990 with persons who have AD and found measurement properties of interrater reliability to be adequate by rater agreement (Volicer, Hurley, Lathi, &Kowall, 1994) and internal consistency to be acceptable by examination of Cronbach’s alpha (Hurley, Volicer, Hanrahan, Houde, & Volicer, 1992; Mahoney, E. K., et al., 1999; Camberg et al., 1999). Our experience shows that research assistants can easily learn to administer the MMSE, are accurate and consistent in its administration to patients following training, and do not upset patients when using it. Because the MMSE is used in so many studies, it is almost incumbent on researchers to include the MMSE to characterize subjects so that consumers of research have a benchmark of cognitive capacity for comparing results across studies. The MMSE is a brief scale that can be administered to patients who have very different levels of cognitive impairment, from no impairment to being quite impaired. However, once a patient scores “0,” the MMSE does not quantify the cognitive differences that can exist between patients who all score “0.” If it is important to clearly characterize research participants in an AD project, another scale such as the Bedford Alzheimer Nursing Subscale (BANS) (Volicer et al., 1994) does allow additional discrimination of dementia severity for subjects who “bottom” on the MMSE. Because the MMSE does not measure executive function, the MMSE alone should not be used as an enrollment criterion in AD research, but should be combined with additional neuropsychological tests.
The lack of complexity in administration leads to high levels of rater reliability—a very important feature. The MMSE can be scored without need for a calculator or computing scores and the total can be entered onto the patient’s record or data. Despite ease of administration and scoring of the MMSE, reliability checks need to be in place and testers should periodically be observed for accuracy. For example, one item asks the patient to follow a standardized 3-stage command, not three sequential single commands. During rater training for our research projects, we have found that some testers wanted the patient to do well and needed to be corrected for imprecisely administering (giving three sequential commands versus one 3-stage command) and scoring (“It was so close”).
The MMSE is a brief screening test and scores were never intended to be a proxy for AD severity or to provide a cut score rule for determining when an individual has the capacity to provide informed consent. MMSE scores provide a useful and objective assessment for nurses in a wide range of clinical situations.