人生得意须尽欢，到点我是必下班
1064 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
S
elf-administration of medication (SAM) by patients in hospitals is a way of improving autonomy and encouraging consumer participation in health care. With healthcare practices leading to shorter hospital stays and many patients taking different drugs for multiple conditions, it is important that they are well prepared to manage their medications following hospital discharge.1-3
SAM programs have become increasingly popular in hospitals. In an extensive UK survey of pharmacy managers from 150 of 467 possible hospital trusts, 39 (48%) had SAM plans in operation.4 Of the remaining respon
dents, 12 (15%) indicated that their hospital trust intended to set up a SAM program within the next 12 months. National regulatory bodies, such as the Australian Pharmaceutical Advisory Council,5 recommend that suitably skilled patients should be encouraged to administer their own drugs and monitor their responses. Self-administration practices enable patients to manage their medication regimens at home because they facilitate adequate preparation before hospital discharge.6,7 In a recent intervention study, the effectiveness of a SAM program was evaluated using a comparative group, repeatedmeasures design.8 Patients received either nurse-administered (n = 172) or self-administered medications (n = 178). Patients in the self-administered group had significantly
Development and Validation of the Self-Administration of Medication Tool
Elizabeth Manias, Christine J Beanland, Robin G Riley, and Alison M Hutchinson
Pharmacosociology
Author information provided at the end of the text.
BACKGROUND: Consumer participation in planning and implementing health care is actively encouraged as a means of improving patient outcomes. In assessing the ability of patients to self-medicate, health professionals can identify areas in which patients need assistance, education, and intervention to optimize their health outcomes after discharge.
OBJECTIVE: To develop and validate a tool to quantify the ability of patients to administer their regularly scheduled medications while they are hospitalized.
METHODS: Past research enabled us to develop the Self-Administration of Medication (SAM) tool. Using a Delphi technique of 3 rounds, a panel of expert health professionals established the content validity of the tool. For determining level of agreement in using the SAM tool, 56 patients were selected; for each patient, 2 randomly selected nurses completed an assessment. Construct validity and internal consistency were examined by testing the tool in 50 patients and comparing with other validated scales.
RESULTS: The 29-item SAM tool had high content validity scores for clarity, representation, and comprehensiveness, with content validity index values ranging from 0.95–1.0. In testing the level of agreement between 2 nurses, out of 43 valid cases, 95.3% of nurses overwhelmingly agreed about the patients’ competence to self-administer their drugs. The intraclass correlation coefficient was 0.819 (95% CI 0.666 to 0.902). Internal consistency for the SAM tool was high, with a Cronbach’s alpha of 0.899. A moderate to strong correlation was obtained when comparing the SAM tool with other validated measures.
CONCLUSIONS: The SAM tool is valid and reliable for quantifying patients’ ability to manage their regularly scheduled medications in the hospital setting.
KEY WORDS: competence, medication administration, medication knowledge, self-administration.
Ann Pharmacother 2006;40:1064-73.
Published Online, 30 May 2006, www.theannals.com, DOI 10.1345/aph.1G677
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
higher drug knowledge scores at discharge and at 2, 6, and 16 weeks after discharge. They also had significantly fewer errors and medication-related problems over time compared with the nurse-administered group. Empirical evidence exists for instruments simulating patients’ management of medications, usually in community settings. These instruments can be divided into the following 3 types: functional assessment of tasks, medication knowledge, and understanding about drugs. Task assessment involves the ability to read medication labels, the effects of color vision and short-term memory, manipulation of child-resistant containers, and label interpretation.9-14 In particular, the Medication Management Test measures high level adaptive functioning in persons with early dementia, based on a structured task for simulating calculations and manipulating drug administration.12 The Hopkins Medication Schedule tests an individual’s ability to complete a daily schedule for taking medications and fill in compartments of a daily pill box.14 Assessment of the patient’s knowledge is the focus of the MedTake Test, which evaluates understanding of dosage, indication, food or water coingestion, and regimen schedule,15 and of the Drug Regimen Unassisted Grading Scale, which examines understanding of identification, access, dosage, and timing of medications.16,17 Assessment of understanding involves testing patients’ capacity to adhere to a treatment regimen presented as a series of 3 graded scenarios.18 A tool that describes the ability to manage medications must, therefore, consider the complex relationships among cognitive function, comprehension, and adherence to drug therapy.19
While evidence exists for simulating aspects of medication management, there is little specific information on determining whether a patient is able to administer medications while in the hospital. Valid assessment tools provide valuable data that can be used to foresee the success of a SAM program and the means by which health professionals can confidently achieve consistency in care delivery. Determining patients’ ability to self-medicate has been largely an intuitive decision, without the use of a validated tool. General measures used have included patient orientation to time and place,20 an intent to return to community living,20,21 the ability to speak English,20 the presence of good eyesight,20,21 and the ability to provide informed consent.20,21 Measures relating to patient health have included the presence of a medically stable condition20 and the absence of drug and/or alcohol abuse.21
Specific measures have included the patient’s capacity to obtain and record drug information,21,22 knowledge of self-administration procedures,22 motivation to self-medicate,21 and the complexity of the treatment regimen.23,24
Clearly, a number of factors influence patients’ ability to self-administer their drugs in the hospital, including physical and mental capability, knowledge about medication, experience with self-medication, and willingness to partici
pate in the activity.7 A tool that includes these factors would be useful in predicting whether patients could selfadminister effectively while in the hospital and in identifying areas in which they might require support or education to manage their medications safely following discharge. No validated tool to examine the interrelationships of factors that influence a patient’s ability to self-medicate has been identified. The objective of this study was to develop and test a comprehensive tool to use in determining the ability of patients to self-administer medications in the acute-care setting.
Methods
Using data collected from individual patient interviews, focus groups with health professionals, and a literature review,7 we developed a SAM tool comprising a 29 item scaled instrument (Appendix I). Psychometric properties were determined according to the following 4 areas: content validity, agreement, internal consistency, and construct validity. A panel of experts was used to establish content validity of the tool. Employing a modified Delphi technique of 3 rounds conducted by email, panel members critically evaluated drafts of the tool until it was deemed adequately refined for pilot testing in the clinical setting. The panel comprised 10 individuals from hospitals, universities, or communities, each of whom had specific expertise in tool development, medication management, risk management, or patient behavior. Members were identified from our knowledge of relevant activities completed or research previously disseminated by the experts. Over a 6 month period, we met on 10 occasions to discuss issues raised by experts and refine the tool. Based on findings from pilot testing, we determined that the tool took no longer than approximately 5 minutes to complete. We prevented the Delphi process facilitator from introducing bias in the outcomes of the correspondence sent by expert panel members. This potential problem was addressed by providing the facilitator with concise instructions about how to calculate content validity and by restricting the facilitator’s involvement in further discussions with panel members. Content validity, at both the item and the instrument levels, was calculated using a 4 point Likert scale, with 1 indicating lack of agreement and 4 indicating excellent agreement with the validity of a specific item.25 The expert panel were asked to examine each item to determine whether it was representative, clear, and comprehensive. A rating of 3 or 4 was considered to be good agreement for an item’s validity. Content validity for the entire instrument was determined using the Content Validity Index (CVI).25 It has been suggested that 83% agreement at the item and instrument levels is required for acceptable values of content validity. To determine the level of agreement for using the SAM tool, 56 patients were randomly selected, and 2 randomly chosen nurses completed an assessment for a particular patient. A random number computer program was used in selecting participating patients and nurses to eliminate any researcher bias. All nurses received prior training in use of the tool. Nurses were asked to independently assess the same patient using the SAM tool and document their responses. They were not to communicate with each other about the activity, and the research assistant was in attendance to ensure their compliance. Therefore, it is unlikely that the first assessment had any effect on the second. The 2 assessments were completed within one hour to ensure that the patient’s condition did not change. Agreement between 2 randomly selected nurses for the population group of 56 patients was determined using a graphical method.26 All patients and nurses were recruited from 3 general medical wards of a private, nonprofit hospital. Criteria for patient inclusion were age of more than 18
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1065 www.theannals.com
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
years; the ability to read, write, and understand English; and having a prescription for regularly administered drug therapy. Patients or their nextof-kin were required to provide informed, written consent. By plotting the differences in the total SAM results against their mean and examining the scatter diagram and absolute mean difference obtained, it was possible to make interpretations about the nature of agreement. The intraclass correlation coefficient was also calculated, using the 2 way random-effects model, where both people effects and measure effects are random. Construct validity testing involved determining the extent to which the tool related to 5 other validated tests. The tool’s internal consistency was also analyzed. For these analyses, the SAM tool was administered to a second group of 50 patients in the 3 medical wards, together with the Morisky Adherence Scale (MAS),27 the Mini-Mental State Examination (MMSE),28 a measure of functional independence (the FIM instrument),29 the Chronic Disease Self-Efficacy Scale (CDSES),30 and the Arthritis Impact Measurement Scale: Dexterity Subscale (AIMS-DS).31
These previously validated instruments were selected because they assessed an aspect of patient self-care, including cognition, ability to perform daily activities, knowledge of healthcare needs, and capacity to open containers. To explore the strength of the relationship between total SAM scores and summated scores obtained from existing validated tools, Pearson’s Product–Moment Correlation Coefficients (r) were calculated. To calculate confidence intervals, Fisher’s formula was used to convert r values to new standard z scores.32 The z value confidence intervals were then transformed into the corresponding r value 95% confidence intervals. A 2 sided p value less than 0.05 was considered statistically significant. Cronbach’s alpha was calculated for the subscale and total SAM scores and for summated scores for existing validated tools for determination of internal consistency. The study was reviewed and approved by the institutional review board of the hospital in which the study was conducted.
Results
The SAM tool developed for testing comprised 2 parts. The first part, comprising 5 items was used to collect demographic data about anticipated discharge destination and responsibility for drug administration following discharge. It was also used to determine the patient’s willingness and competence to self-administer regular medications. The second part comprised 24 items that assessed the patient’s ability to self-administer medications in the hospital setting, in terms of 3 subscales of patient behavior or activity, that is, the capability to self-medicate, knowledge about drugs, and experience with self-medication. The panel assigned weightings to the 24 items of the second part, and a total score was calculated, based on the sum of the Likert scores. Therefore, from a possible score of 96, the panel determined that a minimum score of 60 was required for a patient to be considered competent to self-administer drugs while in the hospital. For the 3 rounds in which the Delphi technique was used, response rates of 20%, 40%, and 70%, respectively, were received from the 10 experts on the panel (Table 1). The CVIs varied from 77% to 100% for the 3 rounds. By the third round, the CVIs obtained for the 3 qualifiers (representative, clear, comprehensive) were between 95% and
100% at the instrument level and were, therefore, above the required level of 83%.25
Table 2 indicates the demographic data for the patients (n = 56) involved in the agreement segment and for those (n = 50) involved in the construct validity and internal consistency determinations of the study. All patients spoke English at home except one who spoke Italian. The second group of patients was formally tested for cognitive impairment, using the MMSE. A standardized score of 23 or lower indicates impairment.28 Of the 50 patients tested, 7 (14%) had cognitive impairment. Figure 1 shows the scatter plot of the absolute difference between the SAM scores against the average of the 2 SAM scores obtained to determine agreement between the 2 nurses. Cases were filtered if there were 3 or more missing values. In total, 43 valid cases were obtained. Of those, the panel members agreed that, in 38 cases, patients were perceived as competent (score ≥60) to self-administer; in 2 cases, patients were perceived as not competent (score range 45–50) to self-administer. Given the absolute mean difference of 7.16, in 3 cases there may have been some lack of clarity about whether patients were perceived as competent to self-administer (score range 55–60). Thus, for 93% of the valid cases, nurses overwhelmingly agreed on patients’ competence or lack of competence to self-administer medications. The scatter plot also suggests that there is greater discrepancy with data when nurses scored low on the SAM scale. The intraclass correlation coefficient for the SAM scores between the 2 nurses, using the average measures method, was 0.819 (95% CI 0.666 to 0.902; F[42,42] = 5.520; p < 0.001). Table 3 shows internal consistency scores and correlations for the 5 validated scales and the SAM tool. Of the 24 items in the second part of the SAM tool assessed for corrected item–total correlation, 3 had values less than 0.30. The 3 items and their respective corrected item–total correlations were “uses an interpreter when required” (0.167), “uses an aide-memoire to assist in self-medica
1066 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
E Manias et al.
Table 1. Content Validity Index Results
Responses for 3 Delphi Rounds
Round 1 Round 2 Round 3
Parameter (n = 2), % (n = 4), % (n = 7), %
Representativeness 98.2–100 77.0–99.1 97.5–100 Clarity 98.2–100 91.4–98.3 95.0–100 Comprehensiveness 98.2–100 99.1–100 95.0–100 Minimum expected 98.2–100 99.1–100 95.0–100 score (weighting for each item)a
aMinimum expected scores (or weightings) for all items were added together to obtain the overall minimum score of 60 required for patients to be considered competent to self-medicate.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
tion” (0.238), and “has independently self-administered in hospital” (0.104). However, removal of any of these items did not improve the internal consistency score of 0.899. The total SAM score correlated strongly with scores obtained from the MMSE and the FIM instrument and moderately correlated with the MAS, the CDSES, and the AIM-DS.
Discussion
This study has demonstrated that the SAM tool is useful for assessing patients’ competence to self-administer medications. Depending on patient scores for the various subscales, it is possible to tailor an individualized, graduated program with increasing levels of patient responsibility for self-administration. The success of a SAM program depends largely on patients’ ability and willingness to participate. For the first time, a tool has been developed to critically and accurately identify patients’ changing requirements over the course of their treatment and their competence to self-medicate. As a new tool, the internal consistency coefficient of 0.899 was considered excellent, and omitting specific questions did not significantly alter the questionnaire as a whole. Following 3 rounds of a Delphi strategy, the CVIs ranged from 95% to 100%, well above the required gold standard of 83%.25
One of the strengths of this tool is that it was developed from patient and health professional interviews7 and tested with nurses and patients in the practice setting. The panel that determined content validity was a team of experts from different backgrounds, which helped produce a tool that addressed the interdisciplinary perspectives of pharmacy, nursing, and medicine. Nurses agreed on patients’ competence (or lack thereof) to self-administer in most cases. The findings demonstrated that measurement was quite reliable at the higher end of the scale, but that discrepancy between the 2 nurses increased at lower SAM scores. Hence, repeatability may be further refined with research into the use of the SAM tool on patients with lower levels of competency to provide better discrimination in this area. Strong to moderate correlations between the SAM tool and various validated instruments indicate its ability to take account of many dimensions involved with successful medication administration. In particular, strong correlations were obtained when the SAM tool was compared with the MMSE and FIM instruments. Such results demonstrate the value of assessing cognitive ability and independent conduct of common daily activities in identifying patient competence to perform self-medication in hospital. Moderate correlation was shown with the MAS, suggesting an association between the likelihood that patients will take drugs as prescribed and their competence to selfmedicate. While moderate correlation was obtained with
Development and Validation of the Self-Administration of Medication Tool
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1067 www.theannals.com
Table 2. Demographic Profiles of Patients
Pts. in Internal Pts. in Consistency/ Agreement Construct Segment Validity Segment
Parameter (N = 56), n (%) (N = 50), n (%)
Gender male 27 (48.2) 30 (60) Aids hearing aid 3 (5.4) 5 (10) glasses 43 (76.8) 46 (92) walking frame 0 (0) 1 (2) Diagnosis on admission, based on body system cardiovascular 31 (55.4) 32 (64) respiratory 8 (14.3) 6 (12) musculoskeletal 8 (14.3) 4 (8) neurologic 0 (0) 2 (4) gastrointestinal 6 (10.7) 5 (10) integument 1 (1.8) 1 (2) renal 2 (3.6) 0 (0) Medical conditions (n) 1 7 (12.5) 9 (18) 2–3 15 (26.8) 11 (22) 4–5 18 (32.1) 22 (44) >5 16 (28.6) 8 (16) Discharge destination own home 47 (83.9) 41 (82) family/caregiver’s home 2 (3.6) 5 (10) residential care facility 2 (3.6) 2 (4) other acute care facility 1 (1.8) 2 (4) rehabilitation 4 (7.1) 0 (0) Likely medications management postdischarge self 45 (80.4) 40 (80) with assistance from 8 (14.3) 7 (14) family/caregiver with assistance from 1 (1.8) 1 (2) community pharmacist with assistance from 2 (3.6) 2 (4) other health professional Pt. will require education to manage medication postdischarge yes 32 (57.1) 27 (54) no 20 (35.7) 12 (24) unable to assess at this 4 (7.1) 11 (22) time Pt. will require support to manage medication (eg, opening container) postdischarge yes 13 (23.2) 8 (16) no 39 (69.6) 34 (68) unable to assess at this 4 (7.1) 8 (16) time Agea 73.84 ± 9.56 72.16 ± 11.32 Regularly scheduled drugs 7.77 ± 3.40 7.64 ± 3.10 prescribed in hospital (n)a
Pt.’s confidence in ability 8.69 ± 2.16 8.84 ± 2.86 to self-manage regular medicationsa,b
Pt.’s desire to self-manage 6.08 ± 4.05 3.88 ± 4.17 drug administration in hospitala,b
aMean ± SD. bVisual analog scale of 0–10 used; 0 = not at all competent or does not want to manage; 10 = completely competent or wants to self-manage medication.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
the AIMS-DS, the confidence interval was large. This large spread could have been due to the specific nature of information obtained from the Dexterity Subscale, which focuses only on patients’ capacity to open containers and manipulate fingers. There are many advantages associated with the SAM tool compared with other instruments that simulate patients’ management of medications.9-19 We tested the SAM tool in hospitalized patients, whereas other instruments have been examined primarily in community-dwelling individuals.9,11-17,19 The SAM tool assesses patients’ perceptions of their desire and competence to self-administer drug therapy in hospital, while other tools do not seek information about these perceptions. Most importantly, the SAM tool comprehensively examines the following 3 areas of self-administration: capability of self-medication, knowledge of drugs, and experience with self-medication. Other instruments tend to target only a single dimension of medication management. This study has some limitations. While we attempted to recruit patients of different levels of competency, the majority demonstrated high levels of ability to self-medicate. Additional work is required to determine the ability of the SAM tool to discriminate accurately between patients of different levels of ability. The study was conducted with patients in a private, nonprofit hospital located in a middleclass suburb. The majority of patients came from an An
glo-Saxon background. Patient views about self-medication could have been different for those in public teaching hospitals that are situated in low socioeconomic areas where the cultural mix is generally greater. Difficulty in accessing appropriately trained interpreters precluded us from recruiting patients who did not understand English. As a result, the dominant view related to patients who were born in Australia. Nevertheless, the study sample is similar to population demographics for private hospital admissions in terms of age, gender, and presenting diagnosis.33
Conclusions
The data suggest that the SAM tool is valid and reliable; however, we recommend that additional testing be carried out in various settings to assess its applicability in different patient populations and in patients of varying levels of ability. Further validation of the tool is also warranted in environments where self-administration is actually implemented.
Elizabeth Manias RN MPharm PhD, Associate Professor, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Carlton Victoria Australia Christine J Beanland RN PhD, Private Medicines Consultant, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne Robin G Riley RN PhD, Senior Lecturer, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne
Alison M Hutchinson RN PhD, Postdoctoral Fellow, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne Reprints: Dr. Manias, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Level 1, 723 Swanston St., Carlton Victoria Australia 3053, fax 61 3 9347 4172, emanias@unimelb.edu.au
We thank the Nurses Board of Victoria, which provided a major research grant to support this study. The views expressed in this article do not necessarily represent those of the Nurses Board of Victoria.
The FIM instrument is a trademark of Uniform Data System for Medical Rehabilitation (UDSMR), a division of UB Foundation Activities, Inc. The use of the FIM instrument to collect data for this research study was authorized by and conducted in accordance with the terms of a special purpose license granted to the authors by UDSMR. The patient data collected during the course of this study have not been processed by UDSMR. No implication is intended that such data have been or will be subjected to UDSMR’s standard data processing procedures or that they are otherwise comparable to data processed by UDSMR.
References
1. Kerzman H, Baron-Epel O, Toren O. What do discharged patients know about their medication? Patient Educ Couns 2005;56:276-82. DOI:10.1016/j.pec.2004.02.019 2. Meredith S, Feldman PH, Frey D, et al. Possible medication errors in home healthcare patients. J Am Geriatr Soc 2001;49:719-24. DOI:10.1046/j.1532-5415.2001.49147.x 3. Runciman WB, Roughead EE, Semple SJ, Adams RJ. Adverse drug events and medication errors in Australia. Int J Qual Health Care 2003;15(suppl 1):i49-59. 4. Ansar S, Silverthorne J. Patients’ own drugs and self-administration of medication schemes in the United Kingdom. Int J Pharm Pract 2002:10:R31.
1068 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
E Manias et al.
Figure 1. Scatter plot of the absolute difference between the SAM scores against the average of the 2 SAM scores (n = 43). There are missing values for the SAM scores because some nurses did not enter a response to particular items. SAM = Self-Administration of Medication.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
5. Australian Pharmaceutical Advisory Council. Guiding principles to achieve continuity in medication management. Canberra: Commonwealth of Australia, 2005. 6. Pereles L, Romonko L, Murzyn T, et al. Evaluation of a self-medication program. J Am Geriatr Soc 1996;44:161-5. 7. Manias E, Beanland C, Riley R, Baker L. Self-administration of medication in hospital: patients’ perspectives. J Adv Nurs 2004;46:194-203. 8. Jensen L. Self-administered cardiac medication program evaluation. Can J Cardiovasc Nurs 2003;13:35-44. 9. Hurd PD, Butkovick SL. Compliance problems and the older patient: assessing functional limitations. Drug Intell Clin Pharm 1986;20:228-31. 10. Meyer ME, Schuna AA. Assessment of geriatric patients’ functional ability to take medication. Drug Intell Clin Pharm 1989;23:171-4. 11. Ruscin JM, Semla TP. Assessment of medication management skills in older outpatients. Ann Pharmacother 1996;30:1083-8. 12. Gurland BJ, Cross P, Chen J, et al. A new performance test of adaptive cognitive functioning: the Medication Management (MM) Test. Int J Geriatr Psychiatry 1994;9:875-85. 13. Fulmer T, Gurland B. Evaluating the caregiver’s intervention in the elder’s task performance: capacity versus actual behavior. Int J Geriatr Psychiatry 1997;12:920-5. 14. Carlson MC, Fried LP, Xue QL, Tekwe C, Brandt J. Validation of the Hopkins Medication Schedule to identify difficulties in taking medications. J Gerontol A Biol Sci Med Sci 2005;60:217-23. 15. Raehl CL, Bond CA, Woods T, Patry RA, Sleeper RB. Individualized drug use assessment in the elderly. Pharmacotherapy 2002;22:1239-48. 16. Edelberg HK, Shallenberger E, Wei JY. Medication management capacity in highly functioning community-living older adults: detection of early deficits. J Am Geriatr Soc 1999;47:592-6. 17. Edelberg HK, Shallenberger E, Hausdorff JM, Wei JY. One-year followup of medication management capacity in highly functioning older adults. J Gerontol A Biol Sci Med Sci 2000;55:M550-3. 18. Fitten LJ, Coleman L, Siembieda DW, Yu M, Ganzell S. Assessment of capacity to comply with medication regimens in older patients. J Am Geriatr Soc 1995;43:361-7. 19. Isaac LM, Tamblyn RM. Compliance and cognitive function: a methodological approach to measuring unintentional errors in medication com
pliance in the elderly. McGill-Calgary Drug Research Team. Gerontologist 1993;33:772-81. 20. Mitchell A. Developing and testing a self medication protocol in the acute environment. Aust J Adv Nurs 2000;17:35-41. 21. Jones L, Arthurs GJ, Sturman E, Bellis L. Self-medication in acute surgical wards. J Clin Nurs 1996;5:229-32. 22. Visalli H, Johnstone L, Lazzaro CA, Kasky S. Medication self-administration: an outcome-orientated, consumer-driven program. J Nurs Care Qual 1997;11:16-22. 23. Bailey A, Ferguson E, Voss S. Factors affecting an individual’s ability to administer medication. Home Healthcare Nurs 1995;13:57-63. 24. George J, Phun Y-T, Bailey MJ, Kong DCM, Stewart K. Development and validation of the Medication Regimen Complexity Index. Ann Pharmacother 2004;38:1369-76. Epub 20 Jul 2004. DOI 10.1345/aph.1D479 25. Lynn MR. Determination and quantification of content validity. Nurs Res 1986; 35:382-5. 26. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10. 27. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care 1986;24:67-74. 28. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189-98. 29. Linacre JM, Heinemann AW, Wright BD, Granger CV, Hamilton BB. The structure and stability of the Functional Independence Measure. Arch Phys Med Rehabil 1994;75:127-32. 30. Lorig K, Stewart A, Ritter P, González V, Laurent D, Lynch J. Outcome measures for health education and other health care interventions. Thousand Oaks, CA: Sage Publications, 1996:24-5, 41-5. 31. Meenan RF, Gertman PM, Mason JH. Measuring health status in arthritis. The arthritis impact measurement scales. Arthritis Rheum 1980;23:146-52. 32. Snedecor GW, Cochran WG. Statistical methods. 7th ed. Ames, IA: Iowa State University Press, 1980. 33. Australian Institute of Health and Welfare (AIHW). Australia’s Health. Canberra: AIHW, 2004.
EXTRACTO
TRASFONDO: La participación del consumidor en la planificación e implantación de su cuidado de salud se estimula activamente como un medio para mejorar resultados. Al llevar a cabo un avalúo de la habilidad de los pacientes para automedicarse, los profesionales de la salud pueden identificar áreas en las cuales estos necesitan ayuda, educación e intervención para obtener óptimos resultados en su salud luego de salir del hospital.
OBJETIVO: Desarrollar y validar una herramienta para cuantificar la habilidad de los pacientes para administrarse sus medicamentos regulares en el hospital.
MÉTODOS: Investigaciones pasadas capacitaron a los autores para desarrollar la herramienta AutoAdministración de Medicamentos (SAM, por sus siglas en inglés para Self-Administration of Medication). Usando una técnica Delfi de 3 rondas, un panel experto de profesionales de la salud estableció la validez de contenido de la herramienta. Para determinar el nivel de concordancia al usar la herramienta SAM, se seleccionaron 56 pacientes, y para cada paciente 2 enfermeros seleccionados al azar completaron un avalúo. La validez de construcción y la consistencia interna fueron examinadas al probar la herramienta en 50 pacientes contra otras escalas validadas.
RESULTADOS: La herramienta SAM tuvo puntuaciones altas en validez de contenido para claridad, representación y amplitud con valores índice que flucturaron entre 0.95–1.0. Al probar el nivel de concordancia entre 2 enfermeros, de 43 casos vólidos, 95.3% de los enfermeros concordaron abrumadoramente acerca de la competencia de los pacientes para
Development and Validation of the Self-Administration of Medication Tool
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1069 www.theannals.com
Table 3. Internal Consistency Scores and Pearson Product–Moment Correlation Coefficients for 5 Validated Scales and the SAM Scalea
Internal
Scale Consistency Score
Morisky Adherence Scale 0.860 Mini-Mental State Examination 0.606 FIM instrumentb 0.996 Chronic Disease Self-Efficacy Scale 0.954 Arthritis Impact Measurement Scale: Dexterity Subscale 0.768 SAM scale (total score) 0.899 (24 items) SAM scale (capability to self-medicate) 0.761 (11 items) SAM scale (knowledge of medicines) 0.843 (7 items) SAM scale (experience with self-medication) 0.884 (6 items)
r value (95% CI); Correlation of Scales p value (2-tailed)
Total SAM score and Morisky Adherence Scale 0.434 (0.167 to 0.641); p = 0.002 Total SAM score and Mini-Mental State Examination 0.688 (0.504 to 0.812); p = 0.0001 Total SAM score and FIM instrument 0.636 (0.446 to 0.785); p = 0.0001 Total SAM score and Chronic Disease Self-Efficacy 0.498 (0.225 to 0.699); Scale p = 0.001 Total SAM score and Arthritis Impact Measurement 0.361 (0.086 to 0.585);
Scales: Dexterity Subscale p = 0.012
SAM = Self-Administration of Medication. aN = 50. bThe FIM instrument. ©1997 Uniform Data System for Medical Rehabilitation (UDSMR). Used with permission.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
1070 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
E Manias et al.
Appendix I. Patient Self-Administration of Medication in the Acute Care Setting
VAS = visual analog scale.
(continued on page 1071)
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
Development and Validation of the Self-Administration of Medication Tool
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1071 www.theannals.com
autoadministrarse sus medicamentos. El coeficiente de correlación intraclases fue 0.819 (95% CI 0.666 y 0.902). La consistencia interna para la herramienta SAM fue alta con un alfa de Cronbach de 0.899. Al comparar la herramienta SAM con otras medidas validadas se obtuvo una correlación de moderada a fuerte.
CONCLUSIONES: La herramienta SAM es válida y confiable para cuantificar la habilidad de los pacientes para manejar sus propios medicamentos en el escenario de hospital.
Ana E Velez
RÉSUMÉ
MISE EN CONTEXTE: La participation des consommateurs dans la planification et l’implantation des soins de santé est fortement encouragée comme un moyen d’améliorer les soins de santé. En évaluant les habiletés pour l’auto-médication, les professionnels de la santé peuvent identifier les aspects pour lesquels les patients nécessitent de l’aide, de l’éducation ou des interventions afin d’optimiser les soins après le congé de l’hôpital.
OBJECTIF: Développer et valider un outil destiné à quantifier l’habileté des patients à s’administrer eux-mêmes leurs médicaments réguliers à l’hôpital.
MÉTHODES: Des recherches antérieures ont permis aux auteurs de développer l’outil d’auto-administration de médicaments (AAM). En utilisant une technique Delphi de 3 sessions, un comité d’experts de professionnels de la santé a validé l’outil. Afin de déterminer le niveau d’accord de l’AAM, 56 patients ont été sélectionnés et pour chacun d’entre eux, 2 infirmières choisies au hasard ont complété une évaluation. La validité de construction et la cohérence interne ont été vérifiées en comparant l’outil à d’autres échelles validées chez 50 patients.
RÉSULTATS: L’outil AAM a obtenu des scores élevés de validité dans le contenu pour la clarité, la représentation et l’ensemble avec des valeurs variant de 0.95 à 1.0. En testant l’accord entre les 2 infirmières dans 43 cas valides, 95.3% des infirmières ont corroboré la capacité des patients à s’administrer les médicaments eux-mêmes. Le coefficient de corrélation intraclasse était de 0.819 (intervalle de confiance à 95% de 0.666 à 0.902). La cohérence interne était élevée avec un alpha de Cronbach de 0.899. Une corrélation de modérée à élevée a été obtenue dans la comparaison avec d’autres outils validés.
CONCLUSIONS: L’outil AAM est valide et fiable pour quantifier l’habileté des patients à gérer leur propre médication dans un milieu hospitalier.
Nicolas Paquette-Lamontagne
Appendix I. Patient Self-Administration of Medication in the Acute Care Setting (continued)
VAS = visual analog scale.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
1072 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
E Manias et al.
Appendix II. Self-Administration of Medication Assessment Tool
(continued on page 1073)
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
Development and Validation of the Self-Administration of Medication Tool
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1073 www.theannals.com
Appendix II. Self-Administration of Medication Assessment Tool (continued)
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
S
elf-administration of medication (SAM) by patients in hospitals is a way of improving autonomy and encouraging consumer participation in health care. With healthcare practices leading to shorter hospital stays and many patients taking different drugs for multiple conditions, it is important that they are well prepared to manage their medications following hospital discharge.1-3
SAM programs have become increasingly popular in hospitals. In an extensive UK survey of pharmacy managers from 150 of 467 possible hospital trusts, 39 (48%) had SAM plans in operation.4 Of the remaining respon
dents, 12 (15%) indicated that their hospital trust intended to set up a SAM program within the next 12 months. National regulatory bodies, such as the Australian Pharmaceutical Advisory Council,5 recommend that suitably skilled patients should be encouraged to administer their own drugs and monitor their responses. Self-administration practices enable patients to manage their medication regimens at home because they facilitate adequate preparation before hospital discharge.6,7 In a recent intervention study, the effectiveness of a SAM program was evaluated using a comparative group, repeatedmeasures design.8 Patients received either nurse-administered (n = 172) or self-administered medications (n = 178). Patients in the self-administered group had significantly
Development and Validation of the Self-Administration of Medication Tool
Elizabeth Manias, Christine J Beanland, Robin G Riley, and Alison M Hutchinson
Pharmacosociology
Author information provided at the end of the text.
BACKGROUND: Consumer participation in planning and implementing health care is actively encouraged as a means of improving patient outcomes. In assessing the ability of patients to self-medicate, health professionals can identify areas in which patients need assistance, education, and intervention to optimize their health outcomes after discharge.
OBJECTIVE: To develop and validate a tool to quantify the ability of patients to administer their regularly scheduled medications while they are hospitalized.
METHODS: Past research enabled us to develop the Self-Administration of Medication (SAM) tool. Using a Delphi technique of 3 rounds, a panel of expert health professionals established the content validity of the tool. For determining level of agreement in using the SAM tool, 56 patients were selected; for each patient, 2 randomly selected nurses completed an assessment. Construct validity and internal consistency were examined by testing the tool in 50 patients and comparing with other validated scales.
RESULTS: The 29-item SAM tool had high content validity scores for clarity, representation, and comprehensiveness, with content validity index values ranging from 0.95–1.0. In testing the level of agreement between 2 nurses, out of 43 valid cases, 95.3% of nurses overwhelmingly agreed about the patients’ competence to self-administer their drugs. The intraclass correlation coefficient was 0.819 (95% CI 0.666 to 0.902). Internal consistency for the SAM tool was high, with a Cronbach’s alpha of 0.899. A moderate to strong correlation was obtained when comparing the SAM tool with other validated measures.
CONCLUSIONS: The SAM tool is valid and reliable for quantifying patients’ ability to manage their regularly scheduled medications in the hospital setting.
KEY WORDS: competence, medication administration, medication knowledge, self-administration.
Ann Pharmacother 2006;40:1064-73.
Published Online, 30 May 2006, www.theannals.com, DOI 10.1345/aph.1G677
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
higher drug knowledge scores at discharge and at 2, 6, and 16 weeks after discharge. They also had significantly fewer errors and medication-related problems over time compared with the nurse-administered group. Empirical evidence exists for instruments simulating patients’ management of medications, usually in community settings. These instruments can be divided into the following 3 types: functional assessment of tasks, medication knowledge, and understanding about drugs. Task assessment involves the ability to read medication labels, the effects of color vision and short-term memory, manipulation of child-resistant containers, and label interpretation.9-14 In particular, the Medication Management Test measures high level adaptive functioning in persons with early dementia, based on a structured task for simulating calculations and manipulating drug administration.12 The Hopkins Medication Schedule tests an individual’s ability to complete a daily schedule for taking medications and fill in compartments of a daily pill box.14 Assessment of the patient’s knowledge is the focus of the MedTake Test, which evaluates understanding of dosage, indication, food or water coingestion, and regimen schedule,15 and of the Drug Regimen Unassisted Grading Scale, which examines understanding of identification, access, dosage, and timing of medications.16,17 Assessment of understanding involves testing patients’ capacity to adhere to a treatment regimen presented as a series of 3 graded scenarios.18 A tool that describes the ability to manage medications must, therefore, consider the complex relationships among cognitive function, comprehension, and adherence to drug therapy.19
While evidence exists for simulating aspects of medication management, there is little specific information on determining whether a patient is able to administer medications while in the hospital. Valid assessment tools provide valuable data that can be used to foresee the success of a SAM program and the means by which health professionals can confidently achieve consistency in care delivery. Determining patients’ ability to self-medicate has been largely an intuitive decision, without the use of a validated tool. General measures used have included patient orientation to time and place,20 an intent to return to community living,20,21 the ability to speak English,20 the presence of good eyesight,20,21 and the ability to provide informed consent.20,21 Measures relating to patient health have included the presence of a medically stable condition20 and the absence of drug and/or alcohol abuse.21
Specific measures have included the patient’s capacity to obtain and record drug information,21,22 knowledge of self-administration procedures,22 motivation to self-medicate,21 and the complexity of the treatment regimen.23,24
Clearly, a number of factors influence patients’ ability to self-administer their drugs in the hospital, including physical and mental capability, knowledge about medication, experience with self-medication, and willingness to partici
pate in the activity.7 A tool that includes these factors would be useful in predicting whether patients could selfadminister effectively while in the hospital and in identifying areas in which they might require support or education to manage their medications safely following discharge. No validated tool to examine the interrelationships of factors that influence a patient’s ability to self-medicate has been identified. The objective of this study was to develop and test a comprehensive tool to use in determining the ability of patients to self-administer medications in the acute-care setting.
Methods
Using data collected from individual patient interviews, focus groups with health professionals, and a literature review,7 we developed a SAM tool comprising a 29 item scaled instrument (Appendix I). Psychometric properties were determined according to the following 4 areas: content validity, agreement, internal consistency, and construct validity. A panel of experts was used to establish content validity of the tool. Employing a modified Delphi technique of 3 rounds conducted by email, panel members critically evaluated drafts of the tool until it was deemed adequately refined for pilot testing in the clinical setting. The panel comprised 10 individuals from hospitals, universities, or communities, each of whom had specific expertise in tool development, medication management, risk management, or patient behavior. Members were identified from our knowledge of relevant activities completed or research previously disseminated by the experts. Over a 6 month period, we met on 10 occasions to discuss issues raised by experts and refine the tool. Based on findings from pilot testing, we determined that the tool took no longer than approximately 5 minutes to complete. We prevented the Delphi process facilitator from introducing bias in the outcomes of the correspondence sent by expert panel members. This potential problem was addressed by providing the facilitator with concise instructions about how to calculate content validity and by restricting the facilitator’s involvement in further discussions with panel members. Content validity, at both the item and the instrument levels, was calculated using a 4 point Likert scale, with 1 indicating lack of agreement and 4 indicating excellent agreement with the validity of a specific item.25 The expert panel were asked to examine each item to determine whether it was representative, clear, and comprehensive. A rating of 3 or 4 was considered to be good agreement for an item’s validity. Content validity for the entire instrument was determined using the Content Validity Index (CVI).25 It has been suggested that 83% agreement at the item and instrument levels is required for acceptable values of content validity. To determine the level of agreement for using the SAM tool, 56 patients were randomly selected, and 2 randomly chosen nurses completed an assessment for a particular patient. A random number computer program was used in selecting participating patients and nurses to eliminate any researcher bias. All nurses received prior training in use of the tool. Nurses were asked to independently assess the same patient using the SAM tool and document their responses. They were not to communicate with each other about the activity, and the research assistant was in attendance to ensure their compliance. Therefore, it is unlikely that the first assessment had any effect on the second. The 2 assessments were completed within one hour to ensure that the patient’s condition did not change. Agreement between 2 randomly selected nurses for the population group of 56 patients was determined using a graphical method.26 All patients and nurses were recruited from 3 general medical wards of a private, nonprofit hospital. Criteria for patient inclusion were age of more than 18
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1065 www.theannals.com
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
years; the ability to read, write, and understand English; and having a prescription for regularly administered drug therapy. Patients or their nextof-kin were required to provide informed, written consent. By plotting the differences in the total SAM results against their mean and examining the scatter diagram and absolute mean difference obtained, it was possible to make interpretations about the nature of agreement. The intraclass correlation coefficient was also calculated, using the 2 way random-effects model, where both people effects and measure effects are random. Construct validity testing involved determining the extent to which the tool related to 5 other validated tests. The tool’s internal consistency was also analyzed. For these analyses, the SAM tool was administered to a second group of 50 patients in the 3 medical wards, together with the Morisky Adherence Scale (MAS),27 the Mini-Mental State Examination (MMSE),28 a measure of functional independence (the FIM instrument),29 the Chronic Disease Self-Efficacy Scale (CDSES),30 and the Arthritis Impact Measurement Scale: Dexterity Subscale (AIMS-DS).31
These previously validated instruments were selected because they assessed an aspect of patient self-care, including cognition, ability to perform daily activities, knowledge of healthcare needs, and capacity to open containers. To explore the strength of the relationship between total SAM scores and summated scores obtained from existing validated tools, Pearson’s Product–Moment Correlation Coefficients (r) were calculated. To calculate confidence intervals, Fisher’s formula was used to convert r values to new standard z scores.32 The z value confidence intervals were then transformed into the corresponding r value 95% confidence intervals. A 2 sided p value less than 0.05 was considered statistically significant. Cronbach’s alpha was calculated for the subscale and total SAM scores and for summated scores for existing validated tools for determination of internal consistency. The study was reviewed and approved by the institutional review board of the hospital in which the study was conducted.
Results
The SAM tool developed for testing comprised 2 parts. The first part, comprising 5 items was used to collect demographic data about anticipated discharge destination and responsibility for drug administration following discharge. It was also used to determine the patient’s willingness and competence to self-administer regular medications. The second part comprised 24 items that assessed the patient’s ability to self-administer medications in the hospital setting, in terms of 3 subscales of patient behavior or activity, that is, the capability to self-medicate, knowledge about drugs, and experience with self-medication. The panel assigned weightings to the 24 items of the second part, and a total score was calculated, based on the sum of the Likert scores. Therefore, from a possible score of 96, the panel determined that a minimum score of 60 was required for a patient to be considered competent to self-administer drugs while in the hospital. For the 3 rounds in which the Delphi technique was used, response rates of 20%, 40%, and 70%, respectively, were received from the 10 experts on the panel (Table 1). The CVIs varied from 77% to 100% for the 3 rounds. By the third round, the CVIs obtained for the 3 qualifiers (representative, clear, comprehensive) were between 95% and
100% at the instrument level and were, therefore, above the required level of 83%.25
Table 2 indicates the demographic data for the patients (n = 56) involved in the agreement segment and for those (n = 50) involved in the construct validity and internal consistency determinations of the study. All patients spoke English at home except one who spoke Italian. The second group of patients was formally tested for cognitive impairment, using the MMSE. A standardized score of 23 or lower indicates impairment.28 Of the 50 patients tested, 7 (14%) had cognitive impairment. Figure 1 shows the scatter plot of the absolute difference between the SAM scores against the average of the 2 SAM scores obtained to determine agreement between the 2 nurses. Cases were filtered if there were 3 or more missing values. In total, 43 valid cases were obtained. Of those, the panel members agreed that, in 38 cases, patients were perceived as competent (score ≥60) to self-administer; in 2 cases, patients were perceived as not competent (score range 45–50) to self-administer. Given the absolute mean difference of 7.16, in 3 cases there may have been some lack of clarity about whether patients were perceived as competent to self-administer (score range 55–60). Thus, for 93% of the valid cases, nurses overwhelmingly agreed on patients’ competence or lack of competence to self-administer medications. The scatter plot also suggests that there is greater discrepancy with data when nurses scored low on the SAM scale. The intraclass correlation coefficient for the SAM scores between the 2 nurses, using the average measures method, was 0.819 (95% CI 0.666 to 0.902; F[42,42] = 5.520; p < 0.001). Table 3 shows internal consistency scores and correlations for the 5 validated scales and the SAM tool. Of the 24 items in the second part of the SAM tool assessed for corrected item–total correlation, 3 had values less than 0.30. The 3 items and their respective corrected item–total correlations were “uses an interpreter when required” (0.167), “uses an aide-memoire to assist in self-medica
1066 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
E Manias et al.
Table 1. Content Validity Index Results
Responses for 3 Delphi Rounds
Round 1 Round 2 Round 3
Parameter (n = 2), % (n = 4), % (n = 7), %
Representativeness 98.2–100 77.0–99.1 97.5–100 Clarity 98.2–100 91.4–98.3 95.0–100 Comprehensiveness 98.2–100 99.1–100 95.0–100 Minimum expected 98.2–100 99.1–100 95.0–100 score (weighting for each item)a
aMinimum expected scores (or weightings) for all items were added together to obtain the overall minimum score of 60 required for patients to be considered competent to self-medicate.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
tion” (0.238), and “has independently self-administered in hospital” (0.104). However, removal of any of these items did not improve the internal consistency score of 0.899. The total SAM score correlated strongly with scores obtained from the MMSE and the FIM instrument and moderately correlated with the MAS, the CDSES, and the AIM-DS.
Discussion
This study has demonstrated that the SAM tool is useful for assessing patients’ competence to self-administer medications. Depending on patient scores for the various subscales, it is possible to tailor an individualized, graduated program with increasing levels of patient responsibility for self-administration. The success of a SAM program depends largely on patients’ ability and willingness to participate. For the first time, a tool has been developed to critically and accurately identify patients’ changing requirements over the course of their treatment and their competence to self-medicate. As a new tool, the internal consistency coefficient of 0.899 was considered excellent, and omitting specific questions did not significantly alter the questionnaire as a whole. Following 3 rounds of a Delphi strategy, the CVIs ranged from 95% to 100%, well above the required gold standard of 83%.25
One of the strengths of this tool is that it was developed from patient and health professional interviews7 and tested with nurses and patients in the practice setting. The panel that determined content validity was a team of experts from different backgrounds, which helped produce a tool that addressed the interdisciplinary perspectives of pharmacy, nursing, and medicine. Nurses agreed on patients’ competence (or lack thereof) to self-administer in most cases. The findings demonstrated that measurement was quite reliable at the higher end of the scale, but that discrepancy between the 2 nurses increased at lower SAM scores. Hence, repeatability may be further refined with research into the use of the SAM tool on patients with lower levels of competency to provide better discrimination in this area. Strong to moderate correlations between the SAM tool and various validated instruments indicate its ability to take account of many dimensions involved with successful medication administration. In particular, strong correlations were obtained when the SAM tool was compared with the MMSE and FIM instruments. Such results demonstrate the value of assessing cognitive ability and independent conduct of common daily activities in identifying patient competence to perform self-medication in hospital. Moderate correlation was shown with the MAS, suggesting an association between the likelihood that patients will take drugs as prescribed and their competence to selfmedicate. While moderate correlation was obtained with
Development and Validation of the Self-Administration of Medication Tool
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1067 www.theannals.com
Table 2. Demographic Profiles of Patients
Pts. in Internal Pts. in Consistency/ Agreement Construct Segment Validity Segment
Parameter (N = 56), n (%) (N = 50), n (%)
Gender male 27 (48.2) 30 (60) Aids hearing aid 3 (5.4) 5 (10) glasses 43 (76.8) 46 (92) walking frame 0 (0) 1 (2) Diagnosis on admission, based on body system cardiovascular 31 (55.4) 32 (64) respiratory 8 (14.3) 6 (12) musculoskeletal 8 (14.3) 4 (8) neurologic 0 (0) 2 (4) gastrointestinal 6 (10.7) 5 (10) integument 1 (1.8) 1 (2) renal 2 (3.6) 0 (0) Medical conditions (n) 1 7 (12.5) 9 (18) 2–3 15 (26.8) 11 (22) 4–5 18 (32.1) 22 (44) >5 16 (28.6) 8 (16) Discharge destination own home 47 (83.9) 41 (82) family/caregiver’s home 2 (3.6) 5 (10) residential care facility 2 (3.6) 2 (4) other acute care facility 1 (1.8) 2 (4) rehabilitation 4 (7.1) 0 (0) Likely medications management postdischarge self 45 (80.4) 40 (80) with assistance from 8 (14.3) 7 (14) family/caregiver with assistance from 1 (1.8) 1 (2) community pharmacist with assistance from 2 (3.6) 2 (4) other health professional Pt. will require education to manage medication postdischarge yes 32 (57.1) 27 (54) no 20 (35.7) 12 (24) unable to assess at this 4 (7.1) 11 (22) time Pt. will require support to manage medication (eg, opening container) postdischarge yes 13 (23.2) 8 (16) no 39 (69.6) 34 (68) unable to assess at this 4 (7.1) 8 (16) time Agea 73.84 ± 9.56 72.16 ± 11.32 Regularly scheduled drugs 7.77 ± 3.40 7.64 ± 3.10 prescribed in hospital (n)a
Pt.’s confidence in ability 8.69 ± 2.16 8.84 ± 2.86 to self-manage regular medicationsa,b
Pt.’s desire to self-manage 6.08 ± 4.05 3.88 ± 4.17 drug administration in hospitala,b
aMean ± SD. bVisual analog scale of 0–10 used; 0 = not at all competent or does not want to manage; 10 = completely competent or wants to self-manage medication.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
the AIMS-DS, the confidence interval was large. This large spread could have been due to the specific nature of information obtained from the Dexterity Subscale, which focuses only on patients’ capacity to open containers and manipulate fingers. There are many advantages associated with the SAM tool compared with other instruments that simulate patients’ management of medications.9-19 We tested the SAM tool in hospitalized patients, whereas other instruments have been examined primarily in community-dwelling individuals.9,11-17,19 The SAM tool assesses patients’ perceptions of their desire and competence to self-administer drug therapy in hospital, while other tools do not seek information about these perceptions. Most importantly, the SAM tool comprehensively examines the following 3 areas of self-administration: capability of self-medication, knowledge of drugs, and experience with self-medication. Other instruments tend to target only a single dimension of medication management. This study has some limitations. While we attempted to recruit patients of different levels of competency, the majority demonstrated high levels of ability to self-medicate. Additional work is required to determine the ability of the SAM tool to discriminate accurately between patients of different levels of ability. The study was conducted with patients in a private, nonprofit hospital located in a middleclass suburb. The majority of patients came from an An
glo-Saxon background. Patient views about self-medication could have been different for those in public teaching hospitals that are situated in low socioeconomic areas where the cultural mix is generally greater. Difficulty in accessing appropriately trained interpreters precluded us from recruiting patients who did not understand English. As a result, the dominant view related to patients who were born in Australia. Nevertheless, the study sample is similar to population demographics for private hospital admissions in terms of age, gender, and presenting diagnosis.33
Conclusions
The data suggest that the SAM tool is valid and reliable; however, we recommend that additional testing be carried out in various settings to assess its applicability in different patient populations and in patients of varying levels of ability. Further validation of the tool is also warranted in environments where self-administration is actually implemented.
Elizabeth Manias RN MPharm PhD, Associate Professor, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Carlton Victoria Australia Christine J Beanland RN PhD, Private Medicines Consultant, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne Robin G Riley RN PhD, Senior Lecturer, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne
Alison M Hutchinson RN PhD, Postdoctoral Fellow, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne Reprints: Dr. Manias, School of Nursing, Faculty of Medicine, Dentistry and Health Sciences, The University of Melbourne, Level 1, 723 Swanston St., Carlton Victoria Australia 3053, fax 61 3 9347 4172, emanias@unimelb.edu.au
We thank the Nurses Board of Victoria, which provided a major research grant to support this study. The views expressed in this article do not necessarily represent those of the Nurses Board of Victoria.
The FIM instrument is a trademark of Uniform Data System for Medical Rehabilitation (UDSMR), a division of UB Foundation Activities, Inc. The use of the FIM instrument to collect data for this research study was authorized by and conducted in accordance with the terms of a special purpose license granted to the authors by UDSMR. The patient data collected during the course of this study have not been processed by UDSMR. No implication is intended that such data have been or will be subjected to UDSMR’s standard data processing procedures or that they are otherwise comparable to data processed by UDSMR.
References
1. Kerzman H, Baron-Epel O, Toren O. What do discharged patients know about their medication? Patient Educ Couns 2005;56:276-82. DOI:10.1016/j.pec.2004.02.019 2. Meredith S, Feldman PH, Frey D, et al. Possible medication errors in home healthcare patients. J Am Geriatr Soc 2001;49:719-24. DOI:10.1046/j.1532-5415.2001.49147.x 3. Runciman WB, Roughead EE, Semple SJ, Adams RJ. Adverse drug events and medication errors in Australia. Int J Qual Health Care 2003;15(suppl 1):i49-59. 4. Ansar S, Silverthorne J. Patients’ own drugs and self-administration of medication schemes in the United Kingdom. Int J Pharm Pract 2002:10:R31.
1068 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
E Manias et al.
Figure 1. Scatter plot of the absolute difference between the SAM scores against the average of the 2 SAM scores (n = 43). There are missing values for the SAM scores because some nurses did not enter a response to particular items. SAM = Self-Administration of Medication.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
5. Australian Pharmaceutical Advisory Council. Guiding principles to achieve continuity in medication management. Canberra: Commonwealth of Australia, 2005. 6. Pereles L, Romonko L, Murzyn T, et al. Evaluation of a self-medication program. J Am Geriatr Soc 1996;44:161-5. 7. Manias E, Beanland C, Riley R, Baker L. Self-administration of medication in hospital: patients’ perspectives. J Adv Nurs 2004;46:194-203. 8. Jensen L. Self-administered cardiac medication program evaluation. Can J Cardiovasc Nurs 2003;13:35-44. 9. Hurd PD, Butkovick SL. Compliance problems and the older patient: assessing functional limitations. Drug Intell Clin Pharm 1986;20:228-31. 10. Meyer ME, Schuna AA. Assessment of geriatric patients’ functional ability to take medication. Drug Intell Clin Pharm 1989;23:171-4. 11. Ruscin JM, Semla TP. Assessment of medication management skills in older outpatients. Ann Pharmacother 1996;30:1083-8. 12. Gurland BJ, Cross P, Chen J, et al. A new performance test of adaptive cognitive functioning: the Medication Management (MM) Test. Int J Geriatr Psychiatry 1994;9:875-85. 13. Fulmer T, Gurland B. Evaluating the caregiver’s intervention in the elder’s task performance: capacity versus actual behavior. Int J Geriatr Psychiatry 1997;12:920-5. 14. Carlson MC, Fried LP, Xue QL, Tekwe C, Brandt J. Validation of the Hopkins Medication Schedule to identify difficulties in taking medications. J Gerontol A Biol Sci Med Sci 2005;60:217-23. 15. Raehl CL, Bond CA, Woods T, Patry RA, Sleeper RB. Individualized drug use assessment in the elderly. Pharmacotherapy 2002;22:1239-48. 16. Edelberg HK, Shallenberger E, Wei JY. Medication management capacity in highly functioning community-living older adults: detection of early deficits. J Am Geriatr Soc 1999;47:592-6. 17. Edelberg HK, Shallenberger E, Hausdorff JM, Wei JY. One-year followup of medication management capacity in highly functioning older adults. J Gerontol A Biol Sci Med Sci 2000;55:M550-3. 18. Fitten LJ, Coleman L, Siembieda DW, Yu M, Ganzell S. Assessment of capacity to comply with medication regimens in older patients. J Am Geriatr Soc 1995;43:361-7. 19. Isaac LM, Tamblyn RM. Compliance and cognitive function: a methodological approach to measuring unintentional errors in medication com
pliance in the elderly. McGill-Calgary Drug Research Team. Gerontologist 1993;33:772-81. 20. Mitchell A. Developing and testing a self medication protocol in the acute environment. Aust J Adv Nurs 2000;17:35-41. 21. Jones L, Arthurs GJ, Sturman E, Bellis L. Self-medication in acute surgical wards. J Clin Nurs 1996;5:229-32. 22. Visalli H, Johnstone L, Lazzaro CA, Kasky S. Medication self-administration: an outcome-orientated, consumer-driven program. J Nurs Care Qual 1997;11:16-22. 23. Bailey A, Ferguson E, Voss S. Factors affecting an individual’s ability to administer medication. Home Healthcare Nurs 1995;13:57-63. 24. George J, Phun Y-T, Bailey MJ, Kong DCM, Stewart K. Development and validation of the Medication Regimen Complexity Index. Ann Pharmacother 2004;38:1369-76. Epub 20 Jul 2004. DOI 10.1345/aph.1D479 25. Lynn MR. Determination and quantification of content validity. Nurs Res 1986; 35:382-5. 26. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10. 27. Morisky DE, Green LW, Levine DM. Concurrent and predictive validity of a self-reported measure of medication adherence. Med Care 1986;24:67-74. 28. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189-98. 29. Linacre JM, Heinemann AW, Wright BD, Granger CV, Hamilton BB. The structure and stability of the Functional Independence Measure. Arch Phys Med Rehabil 1994;75:127-32. 30. Lorig K, Stewart A, Ritter P, González V, Laurent D, Lynch J. Outcome measures for health education and other health care interventions. Thousand Oaks, CA: Sage Publications, 1996:24-5, 41-5. 31. Meenan RF, Gertman PM, Mason JH. Measuring health status in arthritis. The arthritis impact measurement scales. Arthritis Rheum 1980;23:146-52. 32. Snedecor GW, Cochran WG. Statistical methods. 7th ed. Ames, IA: Iowa State University Press, 1980. 33. Australian Institute of Health and Welfare (AIHW). Australia’s Health. Canberra: AIHW, 2004.
EXTRACTO
TRASFONDO: La participación del consumidor en la planificación e implantación de su cuidado de salud se estimula activamente como un medio para mejorar resultados. Al llevar a cabo un avalúo de la habilidad de los pacientes para automedicarse, los profesionales de la salud pueden identificar áreas en las cuales estos necesitan ayuda, educación e intervención para obtener óptimos resultados en su salud luego de salir del hospital.
OBJETIVO: Desarrollar y validar una herramienta para cuantificar la habilidad de los pacientes para administrarse sus medicamentos regulares en el hospital.
MÉTODOS: Investigaciones pasadas capacitaron a los autores para desarrollar la herramienta AutoAdministración de Medicamentos (SAM, por sus siglas en inglés para Self-Administration of Medication). Usando una técnica Delfi de 3 rondas, un panel experto de profesionales de la salud estableció la validez de contenido de la herramienta. Para determinar el nivel de concordancia al usar la herramienta SAM, se seleccionaron 56 pacientes, y para cada paciente 2 enfermeros seleccionados al azar completaron un avalúo. La validez de construcción y la consistencia interna fueron examinadas al probar la herramienta en 50 pacientes contra otras escalas validadas.
RESULTADOS: La herramienta SAM tuvo puntuaciones altas en validez de contenido para claridad, representación y amplitud con valores índice que flucturaron entre 0.95–1.0. Al probar el nivel de concordancia entre 2 enfermeros, de 43 casos vólidos, 95.3% de los enfermeros concordaron abrumadoramente acerca de la competencia de los pacientes para
Development and Validation of the Self-Administration of Medication Tool
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1069 www.theannals.com
Table 3. Internal Consistency Scores and Pearson Product–Moment Correlation Coefficients for 5 Validated Scales and the SAM Scalea
Internal
Scale Consistency Score
Morisky Adherence Scale 0.860 Mini-Mental State Examination 0.606 FIM instrumentb 0.996 Chronic Disease Self-Efficacy Scale 0.954 Arthritis Impact Measurement Scale: Dexterity Subscale 0.768 SAM scale (total score) 0.899 (24 items) SAM scale (capability to self-medicate) 0.761 (11 items) SAM scale (knowledge of medicines) 0.843 (7 items) SAM scale (experience with self-medication) 0.884 (6 items)
r value (95% CI); Correlation of Scales p value (2-tailed)
Total SAM score and Morisky Adherence Scale 0.434 (0.167 to 0.641); p = 0.002 Total SAM score and Mini-Mental State Examination 0.688 (0.504 to 0.812); p = 0.0001 Total SAM score and FIM instrument 0.636 (0.446 to 0.785); p = 0.0001 Total SAM score and Chronic Disease Self-Efficacy 0.498 (0.225 to 0.699); Scale p = 0.001 Total SAM score and Arthritis Impact Measurement 0.361 (0.086 to 0.585);
Scales: Dexterity Subscale p = 0.012
SAM = Self-Administration of Medication. aN = 50. bThe FIM instrument. ©1997 Uniform Data System for Medical Rehabilitation (UDSMR). Used with permission.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
1070 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
E Manias et al.
Appendix I. Patient Self-Administration of Medication in the Acute Care Setting
VAS = visual analog scale.
(continued on page 1071)
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
Development and Validation of the Self-Administration of Medication Tool
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1071 www.theannals.com
autoadministrarse sus medicamentos. El coeficiente de correlación intraclases fue 0.819 (95% CI 0.666 y 0.902). La consistencia interna para la herramienta SAM fue alta con un alfa de Cronbach de 0.899. Al comparar la herramienta SAM con otras medidas validadas se obtuvo una correlación de moderada a fuerte.
CONCLUSIONES: La herramienta SAM es válida y confiable para cuantificar la habilidad de los pacientes para manejar sus propios medicamentos en el escenario de hospital.
Ana E Velez
RÉSUMÉ
MISE EN CONTEXTE: La participation des consommateurs dans la planification et l’implantation des soins de santé est fortement encouragée comme un moyen d’améliorer les soins de santé. En évaluant les habiletés pour l’auto-médication, les professionnels de la santé peuvent identifier les aspects pour lesquels les patients nécessitent de l’aide, de l’éducation ou des interventions afin d’optimiser les soins après le congé de l’hôpital.
OBJECTIF: Développer et valider un outil destiné à quantifier l’habileté des patients à s’administrer eux-mêmes leurs médicaments réguliers à l’hôpital.
MÉTHODES: Des recherches antérieures ont permis aux auteurs de développer l’outil d’auto-administration de médicaments (AAM). En utilisant une technique Delphi de 3 sessions, un comité d’experts de professionnels de la santé a validé l’outil. Afin de déterminer le niveau d’accord de l’AAM, 56 patients ont été sélectionnés et pour chacun d’entre eux, 2 infirmières choisies au hasard ont complété une évaluation. La validité de construction et la cohérence interne ont été vérifiées en comparant l’outil à d’autres échelles validées chez 50 patients.
RÉSULTATS: L’outil AAM a obtenu des scores élevés de validité dans le contenu pour la clarité, la représentation et l’ensemble avec des valeurs variant de 0.95 à 1.0. En testant l’accord entre les 2 infirmières dans 43 cas valides, 95.3% des infirmières ont corroboré la capacité des patients à s’administrer les médicaments eux-mêmes. Le coefficient de corrélation intraclasse était de 0.819 (intervalle de confiance à 95% de 0.666 à 0.902). La cohérence interne était élevée avec un alpha de Cronbach de 0.899. Une corrélation de modérée à élevée a été obtenue dans la comparaison avec d’autres outils validés.
CONCLUSIONS: L’outil AAM est valide et fiable pour quantifier l’habileté des patients à gérer leur propre médication dans un milieu hospitalier.
Nicolas Paquette-Lamontagne
Appendix I. Patient Self-Administration of Medication in the Acute Care Setting (continued)
VAS = visual analog scale.
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
1072 I The Annals of Pharmacotherapy I 2006 June, Volume 40 www.theannals.com
E Manias et al.
Appendix II. Self-Administration of Medication Assessment Tool
(continued on page 1073)
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
Development and Validation of the Self-Administration of Medication Tool
The Annals of Pharmacotherapy I 2006 June, Volume 40 I 1073 www.theannals.com
Appendix II. Self-Administration of Medication Assessment Tool (continued)
Downloaded from aop.sagepub.com at UCSF LIBRARY & CKM on February 16, 2015
当一天和尚撞一天钟
当一天牛马发一天疯
当一天牛马发一天疯
人生得意须尽欢
到点我是必下班
到点我是必下班
上班不摸鱼
脑子有问题
脑子有问题
人生得意须尽欢，到点我是必下班
粤ICP备2023068874号