Secondary Logo

What is the relationship between ED patient anxiety level and patient sureness of medication history?

Shapiro, Marcia MSN, RN; Speroni, Karen Gabel PhD, MHSA, BSN, RN; Edsall, D. Renee MSN, FNP, RN, CCRN; Daniel, Marlon G. MHA, MPH

doi: 10.1097/01.NURSE.0000524772.64461.f0
Department: RESEARCH CORNER
Free

How patients' anxiety levels affect the accuracy of medication reconciliation

At the University of Maryland Shore Medical Center at Easton in Easton, Md., Marcia Shapiro is a clinical nurse in the ED and D. Renee Edsall is an emergency services educator. At the University of Maryland Shore Regional Health in Easton, Md., Karen Gabel Speroni is the chair of the nursing research council and Marlon G. Daniel is a biostatistician.

Research Corner is coordinated by Ruth A. Mooney, PhD, MN, RN-BC, the former nursing research facilitator at Christiana Care Health System in Newark, Del.

The content in this article has received appropriate institutional review board and/or administrative approval for publication.

This study received partial funding through a grant from the Epsilon Zeta Chapter of Sigma Theta Tau. The authors have disclosed no other financial relationships related to this article.

Abstract

Background. The Joint Commission's National Patient Safety Goals mandate patient medication reconciliation. ED nurses obtaining patient medication histories noted variances in those provided. The study's purpose was to evaluate the relationship between self-reported ED patient anxiety and patients' sureness of medication history. The hypothesis was that an inverse relationship would exist between lower patient anxiety and sureness regarding medication history. Methods. Adult patients (N = 150) with non-life-threatening conditions ranked their anxiety level and how sure they were about the accuracy of the medication history they had provided, using a 100-mm visual analogue scale (with 0 = none and 100 = most), at three times (time 1 = baseline/ED entry; time 2 = after ED provider contact; and time 3 = ED disposition). Results. Patients were primarily females (58.7%), Caucasians (68.7%), and high school graduates (49.3%); the mean age was 61.7 years. Significant trends were found in mean anxiety (time 1 = 35.7; time 2 = 31.1; and time 3 = 23.8; P < 0.0001) and being most sure of the accuracy of the medication history over time (time 1 = 90.1; time 2 = 92.6; and time 3 = 95.4; P < 0.0016). Times 1 and 2 were not significant; a significant time 3 result demonstrated lower anxiety and patients being most sure of their medication history. Most patients recalled their medication history differently from that given at baseline; 69.3% recalled a difference at time 2 and 89.3% at time 3 (P < 0.0001). Discussion. Findings demonstrate a significant relationship between patient anxiety (lower anxiety at ED disposition than at ED entry or provider contact) and their level of sureness in the accuracy of their medication history (most sure at ED disposition). ED nurses, ED healthcare providers, pharmacists, or pharmacy technicians should consider rechecking medication history at the time of ED disposition. To support patient safety, nurses who work on receiving inpatient units also share responsibility to ensure the accuracy of the medication history because a complete and accurate listing of medications might not always be possible in the ED setting.

Back to Top | Article Outline

Background

To facilitate patient safety and reduce medication errors in hospitals, one of The Joint Commission's Sentinel Event Alerts addressed using medication reconciliation to prevent errors.1 A process is required to obtain and document a complete listing of the patient's current medications. This reconciliation process should begin at the first point of hospital entry. The 2016 National Patient Safety Goals, which address patient safety in part through safe medication administration, specify using medication safely by recording and passing along correct information about a patient's medication.2 The Institute for Healthcare Improvement describes medication reconciliation as a process to create the most accurate list of the patient's medications, including the drug name, dosage, frequency, and route, and also comparing the list created to the healthcare provider's admission, transfer, and/or discharge order.3 The goal is to provide correct medications for the patient at hospital transition points.

Initial steps in gathering complete and accurate medication histories can be time consuming, incomplete, and prone to error.4-6 Discrepancies increase the potential for moderate to severe adverse drug reactions and patient deaths.7,8 Barriers affecting the accuracy of medication histories obtained include patient illness, patient knowledge, availability of patient medication vials for inspection, and lack of access to community pharmacy records.9 Patients experience higher levels of anxiety during medical emergencies, illness, fearful events, and pain, and these experiences also inhibit cognitive reasoning.10 Research conducted about oncology patients' medication knowledge (drug name, dose, frequency, route, and indication) when presenting to an ED reported that those using a survey medication aid were 6.5 times more likely to have a high level of medication knowledge.11

ED nurses typically document in the electronic medical record (EMR) the patient's medication history (medication name, dose, route, frequency, and last dose). Sources of information include the oral report of the patient or person accompanying the patient, patients' paper or electronic lists, medication containers, and the institution's existing EMR. Medications documented may include prescription drugs, over-the-counter drugs, pertinent vaccines, diagnostic and contrast agents, radioactive medications, respiratory therapy-related medications, parenteral nutrition, I.V. solutions, vitamins, herbals, and nutraceuticals.

Nurses must assimilate both the source and type of information. Nurse researchers working in the ED noted that patients provided a more complete medication history at different points during their ED stay. Patient anxiety was thought to possibly impact patients' ability to provide their medication history accurately. Anxiety is defined as a subjective feeling of apprehension in response to perceived danger.12 Signs and symptoms of anxiety include elevated vital signs, dizziness, diaphoresis, restlessness, feelings of inadequacy, increased tension, fear, worry, difficulty concentrating, inability to solve problems, and decreased cognitive reasoning.1,10,13,14

After recognizing that patients' anxiety level might affect the accuracy of the information they provide, ED nurse researchers wanted to determine the best time to begin obtaining and documenting patients' current medications. At that time, patient anxiety levels were not formally measured in the ED.

A literature review was conducted in ProQuest, the Cochrane Database of Systematic Reviews, and PubMed for the years 2000 to 2015. Search terms used were medication reconciliation, anxiety, patient anxiety, cognitive anxiety effects, emergency department anxiety, and emergency nursing. No research was identified that evaluated the relationship between ED patient anxiety and medication histories.

Two studies informed the design for the research study reported below. In one study comparing two anxiety instruments in patients receiving mechanical ventilation, the Visual Analogue Scale for Anxiety (VAS-A) was less burdensome for research participants than the Spielberger State Anxiety Inventory (SAI).15 The VAS-A was a 100-mm scale, with vertical orientation, and with responses of “not anxious at all” at the bottom, and “the most anxious I have ever been” at the top. In another study, which was a cross-sectional survey, anxiety levels of the person accompanying the patient to the ED were measured at arrival and discharge, using the VAS-A.16 (See Glossary of research terms.) Results showed a significant association between anxiety and satisfaction at the point of discharge: The lower the level of anxiety, the more satisfied they were.

The purpose of this study was to evaluate the relationship between self-reported ED patient anxiety and patients' sureness of their medication history. The hypothesis was that an inverse relationship would exist between patient anxiety and sureness regarding medication history. At the time the research was initiated, hospital policy did not mandate when ED nurses or other healthcare professionals would obtain and document medication history (that is, at baseline/ED entry, after ED provider contact, or at ED disposition), or if a follow-up was required to recheck/confirm history obtained. From an evidence-based practice perspective, nursing administrators could use the results of this research to evaluate whether hospital policy needed to include the best time(s) for patient medication histories to be obtained by ED nurses or other healthcare professionals and whether rechecking or confirming medication histories was needed.

Back to Top | Article Outline

Methods

This nonrandomized, repeated measures study included quantitative and qualitative components. It was conducted in a 32-bed ED in a 132-bed hospital in the Mid-Atlantic region. This hospital has American Nurses Credentialing Center Magnet® recognition and is part of a hospital system consisting of three hospitals.

Nurse researchers included in the study those ED patients who met study eligibility criteria after they provided informed consent. Study subjects met these inclusion criteria:

  • age 18 years or older
  • alert and oriented to person, place, and time
  • acuity levels 2 to 5 (2 = high acuity; 3 = stable acuity with numerous resources; 4 = stable acuity with one resource; 5 = stable acuity with no resources)
  • ability to communicate in the English language.

Patients with cognitive, neurologic, or other impairments affecting their ability to provide medication history and those with life-threatening conditions were excluded.

Back to Top | Article Outline

Study procedures

Study procedures occurred at baseline/ED entry (time 1), after ED provider contact (time 2), and at ED disposition (time 3). At each of these times, nurses asked patients to provide a medication history. They also asked patients to complete a study patient evaluation form to rate current anxiety levels on the 100-mm VAS (0 = not anxious at all; 100 = the most anxious I have ever been) and how sure they were of the accuracy of the medication history provided (0 = not sure at all; 100 = the most sure I have ever been). The baseline medication history, obtained according to hospital policy, included a list of medication names, instructions (dose, route, frequency), and date of last dose.

Nurse researchers also collected data about the following: patient demographics; primary ED triage complaint; acuity level of 2 to 5; medications taken in the ED to alter anxiety; ED length of stay; ED disposition; and qualitative data statements provided by subjects during the study period relating to anxiety or medication histories. To obtain a medication history, nurse researchers asked patients these three questions:

  1. What prescription medications are you taking?
  2. What vitamins and herbs are you taking?
  3. Do you take any over-the-counter medications?
Back to Top | Article Outline

Sample size

A sample size of 130 was determined to have 80% power to have a minimal detectable effect of 0.25 (eta-squared) on the VAS scale. The correlation between patient perception of anxiety and medication reconciliation was assumed to be at least r = 0.5. Due to the dynamic nature of an ED, a 15% rate of loss to follow-up (that is, the patient left the ED or was not available) was assumed at all three time points, and a total sample size of 150 was used. In this study, a total of 152 patients provided informed consent to achieve 150 completed patients, defined as those who completed the patient evaluation form at all three points. Two patients did not complete these study requirements.

Back to Top | Article Outline

Qualitative analysis methods

For subjects' qualitative comments about their anxiety or medication histories, conventional content analysis guided the coding.17 Units of analysis were established by one researcher and then independently coded by two additional researchers. The coding discrepancies identified were discussed among the three researchers to achieve 100% consensus.

Back to Top | Article Outline

Data analysis methods

Means, standard deviation (SD), frequencies, and percentages were used to characterize the data. Repeated measures of analysis of variance were used to model the trends in anxiety and sureness in the medication history provided during patients' ED visits. These variables were analyzed and compared at three points (1 = baseline/ED entry; 2 = after ED provider contact; and 3 = ED disposition). Subset analysis investigated differences in the main variables of interest by looking at whether patients took anxiety or pain medications. Multiple regression was employed to calculate partial correlations between the primary variables (current anxiety levels and how sure patients were about the medication history provided), adjusted for the effects of age, gender, race, and educational level. Tests for normality and other parametric assumptions were conducted for all analytical variables and tests. Data analysis was completed using SAS v 9.3 (SAS Institute, Cary, N.C.).

Back to Top | Article Outline

Results

Of the 150 patients completing the study, most were female (88, 58.7%), Caucasians (103, 68.7%), with an average age of 61.7 years [SD = 17.3], and high school was their highest education level (74, 49.3%). The primary ED triage complaints were cardiac issues (30, 20.0%) followed by digestive problems (29, 19.3%) and trauma (23, 15.3%); all subjects had an acuity level between 2 and 5, with most having an acuity level of 3 (109, 72.7%). Average ED length of stay was 236 minutes [SD = 153 minutes], and most were discharged to home (107, 71.3%). At time 1, patients reported an average of 8.7 [SD = 4.8] medications. Sources for medication information were patients' oral report (73, 49.0%), patients' lists (53, 35.6%), medication bottles (15, 10.1%), and oral report of the person accompanying the patient (7, 4.7%).

Table

Table

Overall, as mean patient anxiety level decreased, mean patient sureness in the accuracy of the medication history increased, indicating that, as hypothesized, level of anxiety was inversely related to sureness in the accuracy of the medication history. (See Patient perceptions.)

Significant trends were found in anxiety level (time 1 = 35.7, time 2 = 31.1, and time 3 = 23.8; P < 0.0001) and being most sure of the accuracy of the medication history (time 1 = 90.1, time 2 = 92.6, and time 3 = 95.4; P < 0.0016). At time 1, very low correlations were found between anxiety and sureness of the accuracy of the medication history, and these were not statistically significant (r = −0.04; overall P = 0.6274). A similar pattern was found at time 2, and these were inversely correlated (r = −1.12; overall P = 0.2540). Patients who received medications in the ED that might affect anxiety level were placed in the “med group” and those who did not were the “no med group.” At time 3, a moderate negative association between anxiety and sureness in medication history was found, which was statistically significant (med group, r = −0.41; P = 0.0088, no med group, r = −0.28; P = 0.0049; r = −0.30; overall P = 0.0003).

See Patient perceptions for the overall analysis by subgroup, including whether patients were in the med group or the no med group at each of the times measured. ED nurses administered to 47 (31.3%) patients a medication that might affect their anxiety level. These included antianxiety medications and pain medications. Analysis of this group of 47 patients demonstrated this trend: As their anxiety level decreased, their sureness in their medication history increased.

Changes in medication history from baseline were also evaluated. Most patients recalled their medication history differently at times 2 and 3 compared with time 1. Specifically, at time 2, 69.3% noted a change to what they had specified at baseline, and at time 3, 89.3% specified a change. These findings were significant (P < 0.0001). Analyses with partial correlation estimates determined associations between their anxiety level and sureness in medication history at each point. It was determined that controlling for demographic values such as age, gender, race, and education would mitigate those effects on the relationship between the two main variables of interest, anxiety level and sureness in medication history. A weak inverse relationship was found at time 1 for both the med and no med groups, and overall. The correlations were not statistically significant. Similar findings were found at time 2, with nonsignificant associations. At time 3, the association between anxiety and sureness in the accuracy of the medication history was moderate (r = −0.41, P = 0.0088) for the med group but slightly weaker for the no med group (r = −0.28, P = 0.0049). Overall, at time 3, a moderate inverse relationship was found between anxiety and sureness in the accuracy of the medication history (r = −0.30, P = 0.0003). At time 3, all of these associations were found to be statistically significant.

Back to Top | Article Outline

Qualitative results

When asked, “How much anxiety do you have now?” at time 1, 18 patients (12.0%) made a qualitative comment about a primary theme of a physical ailment contributing to their anxiety. For example, one patient said, “Having chest pain.” When asked, “How sure are you now that the medication history you provided is accurate?,” 16 patients (10.7%) provided a comment with a primary theme of the patient having too many medications. For example, one patient said, “Have lots of medicines in my car.” Another was 100% sure of medication names but did not know doses.

At time 2, when 11 patients (7.3%) were asked, “How much anxiety do you have now?,” they commented on the primary theme of being anxious. For example, one patient said, “I am still anxious”; another said, “I am most anxious. I want to know what is wrong with me.” When asked, “How sure are you now that the medication history you provided is accurate?,” 10 patients (6.7%) commented with a primary theme of a list or other resource for how best to get information. For example, one patient's medication list was brought in from home by a spouse. Another patient called the pharmacy for the name and dose of cholesterol medication, saying, “I called the pharmacy for accuracy.”

At time 3, 16 patients (10.7%) were asked, “How much anxiety do you have now?” Their comments' primary theme was anxiety relating to being admitted or having to undergo surgery. For example, one patient said, “I am most anxious now. I am getting admitted.” Another said, “Most anxious. Awaiting surgery.” When asked, “How sure are you now that the medication history you provided is accurate?,” 8 patients (5.3%) commented with a primary theme of remembering another medication. One patient said, “I forgot to tell you about [an over-the-counter cold medicine]”; another patient remembered a prescribed migraine medication.

Back to Top | Article Outline

Discussion

It is important to make inpatient medication reconciliation patient-centered, clinically relevant, and implementable.18 According to the procedure of the hospital where this research was conducted, the policy is to ensure safe and effective medication reconciliation for patients across the continuum of care. For the outpatient department and ED, when a patient presents, a list of the patient's current medications is to be obtained and documented. Medications used during the patient's encounter are to be compared with those recorded on the patient's medication history to determine whether the medication is to be continued, continued with modifications, or discontinued based on the patient's clinical status. At the completion of treatment, the patient receives written instructions about whether the treatment provided will impact the use of the current medications listed at the time of presentation.

Maintaining a culture of safety can be problematic when patients' initial presentation medication history is incomplete or inaccurate. Nurses and other healthcare professionals at all patient entry-level settings, such as ED healthcare providers, pharmacists, or pharmacy technicians, can be instrumental in taking initial steps for ensuring processes that ultimately promote patient safety. This includes knowing when patients may be most sure of their medication history, particularly when recalling or providing medication-related information is impacted by anxiety, as in this study of ED patients.

This research study validated an inverse relationship between patient anxiety and sureness in medication history. As anxiety decreased, patients became most sure of the medication history they had provided to the ED nurse. These findings are similar to the study in which there was a significant association between anxiety and satisfaction of people accompanying ED patients; the lower the level of anxiety when leaving the ED, the more satisfied they were.16

Building a recheck or confirmation for patient medication history at ED disposition into the EMR could contribute to a decreased potential for medication errors and/or adverse events. Because this is additional documentation in a complex ED environment, best processes for completing this documentation also need to be examined. With this documentation, for ED patients who become inpatients or who subsequently return to the ED, the EMR should then contain the most accurate medication history information available. To support patient safety, nurses who work on receiving inpatient units also share the responsibility to ensure the accuracy of the medication history because a complete and accurate listing of medications might not always be possible in the ED setting.

Additional research is warranted to further explore the effect of a recheck or confirmation function of patient medication history at ED disposition or on the inpatient unit on rates of near-misses, or actual medication errors and/or adverse events. Additional research is also warranted to further evaluate the effect of anxiety and other variables on patient provision of medication history. These include but are not limited to alterations in the patient's mental status due to medications administered or the nature of the illness or preexisting conditions, a high level of activity or acuity in the ED (which may not allow time for the ED nurse or other healthcare professionals to accurately capture the medications on the EMR at disposition), the acuity of the patient who needs to move rapidly from the ED to the inpatient unit, and the need to move the patient to the inpatient bed rapidly to decompress the ED quickly to accommodate an influx of new ED patients. Last, because this study focused on the ED setting, a similar research design could be applied to other entry settings where nurses or other healthcare professionals are responsible for documenting medication history.

Back to Top | Article Outline

Limitations

The patient sample was derived from a rural hospital setting in the Mid-Atlantic region, so results from other settings or geographic areas may differ. The VAS scale in this study used a horizontal axis; some patients may better understand a vertical axis. Although nurse researchers in this study provided both oral and written instructions for patients for ranking the VAS, in research using the VAS, researchers are advised to confirm patient understanding of the rankings. Other variables that were not measured in this study may have contributed to the medication histories provided by patients.

Back to Top | Article Outline

Implications for healthcare

A standardized process must be used to obtain and document a complete list of patients' current medications. Nurses or other healthcare professionals need to complete this for patients at all entry settings, whether the ED or an outpatient or inpatient setting. Nurses should evaluate whether their facility's policy addresses when to obtain and document patients' current medications. These research findings support ED disposition as the time when patients are most sure of the accuracy of their medication history. Nurses can evaluate best policies and consider whether ED nurses or other healthcare professionals should recheck patient medication histories at ED disposition or upon admission to an inpatient unit. Because these research findings show lower levels of patient anxiety correlate to patients being sure of their medication history, consider the patient's anxiety level when asking about their medication history. Further, best use of informatics systems and EMRs for documentation to address these aspects of care at relevant times can be considered.

Back to Top | Article Outline

Conclusions

The study findings demonstrate a significant relationship between patient anxiety levels (lower anxiety levels at ED disposition than at ED entry or provider contact) and patients being most sure in their medication history (higher levels of sureness at ED disposition). Nurses at all patient entry-level settings can be instrumental in ensuring processes to promote patient safety through medication reconciliation. ED nurses or other healthcare professionals should recheck or confirm medication history at ED disposition as well as during the course of the impatient stay and ensure updates to the medications are documented.

Back to Top | Article Outline

Glossary of research terms

  • Content analysis is a research technique for making replicable and valid inferences from texts (or other meaningful matter) to the contexts of their use.19
  • A cross-sectional survey collects information from a population or representative sample from one point in time.20
  • Effect size is a quantitative measure of the estimated difference between two groups.21
  • Eta-squared is a method used to measure the effect size when conducting analysis of variance (ANOVA).22
  • Frequency is the number of times an event occurs or is observed.23
  • Multiple regression is a statistical technique that measures the relationship between one continuous dependent (outcome) variable and a combination of two or more independent (explanatory) variables.21
  • Mean, also known as the arithmetic mean or average, is the sum of all observations divided by the count of all observations.23
  • Nonrandomized repeated measures design is a study design that follows two or more cohorts of individuals over time, where individuals are not randomly assigned to different treatment groups.24
  • Pvalue, or probability value, is the probability, under the null hypothesis, of obtaining a result equal to or more extreme than the observed value.25
  • Pearson's r, also known as Pearson's correlation coefficient, is the measure of the linear relationship between two continuous variables.23
  • Partial correlation, or adjusted Pearson's correlation, is the measure of the relationship between two variables, while controlling or adjusting for the effect of one or more additional variables.26
  • Standard deviation (SD) is the measure of the average variation (dispersion) of individual values from the overall mean.27
  • Test for normality is an assessment of the data to determine if the data follow a normal (bell-shaped) distribution.28
Back to Top | Article Outline

REFERENCES

1. The Joint Commission. Sentinel Event Alert, Issue 35. Using medication reconciliation to prevent errors. 2006;35. http://www.jointcommission.org/assets/1/18/SEA_35.PDF.
2. The Joint Commission. Hospital National Patient Safety Goals, 2016. http://www.jointcommission.org/assets/1/6/2016_NPSG_HAP_ER.pdf.
3. Institute for Healthcare Improvement. Medication reconciliation to prevent adverse drug events. 2017. http://www.ihi.org/explore/ADEsMedicationReconciliation/Pages/default.aspx.
4. Balon J, Thomas SA. Comparison of hospital admission medication lists with primary care physician and outpatient pharmacy lists. J Nurs Scholarsh. 2011;43(3):292–300.
5. Riley-Lawless K. Family-identified barriers to medication reconciliation. J Spec Pediatr Nurs. 2009;14(2):94–101.
6. Mazer M, DeRoos F, Hollander JE, McCusker C, Peacock N, Perrone J. Medication history taking in emergency department triage is inaccurate and incomplete. Acad Emerg Med. 2011;18(1):102–104.
7. Pippins JR, Gandhi TK, Hamann C, et al Classifying and predicting errors of inpatient medication reconciliation. J Gen Intern Med. 2008;23(9):1414–1422.
8. Riley-Lawless K. Family-identified barriers to medication reconciliation. J Spec Pediatr Nurs. 2009;14(2):94–101.
9. Tam VC, Knowles SR, Cornish PL, Fine N, Marchesano R, Etchells EE. Frequency, type and clinical importance of medication history errors at admission to hospital: a systematic review. CMAJ. 2005;173(5):510–515.
10. Miller CR. Patients with anxiety disorders: a challenge for primary care. AAACN Viewpoint. 2001;22(5):1,19–22.
11. Hwang JP, Holmes HM, Kallen MA, et al Accuracy of reporting current medications by cancer patients presenting to an emergency center. Support Care Cancer. 2010;18(10):1347–1354.
12. Lazarus RS, Folkman S. Stress, Appraisal, and Coping. New York, NY: Springer Publishing Co.; 1984.
13. Ekwall A, Gerdtz M, Manias E. Anxiety as a factor influencing satisfaction with emergency department care: perspectives of accompanying persons. J Clin Nurs. 2009;18(24):3489–3497.
14. Pritchard MJ. Using the hospital anxiety and depression scale in surgical patients. Nurs Stand. 2011;25(34):35–41.
15. Chlan LL. Relationship between two anxiety instruments in patients receiving mechanical ventilatory support. J Adv Nurs. 2004;48(5):493–499.
16. Ekwall A, Gerdtz M, Manias E. Anxiety as a factor influencing satisfaction with emergency department care: perspectives of accompanying persons. J Clin Nurs. 2009;18(24):3489–3497.
17. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15(9):1277–1288.
18. Greenwald JL, Halasyamani L, Greene J, et al Making inpatient medication reconciliation patient centered, clinically relevant and implementable: a consensus statement on key principles and necessary first steps. J Hosp Med. 2010;5(8):477–485.
19. Krippendorff K. Content Analysis: An Introduction to Its Methodology. Thousand Oaks, CA: Sage; 2004.
20. Dawson B, Trapp RG. Basic and Clinical Biostatistics. New York, NY: McGraw-Hill Medical; 2004.
21. Tabachnick BG, Fidell LS, Osterlind SJ. Using Multivariate Statistics. Upper Saddle River, NJ: Pearson; 2014.
22. Murphy KR, Myors B, Wolach A. Statistical Power Analysis: A Simple and General Model for Traditional and Modern Hypothesis Tests. 4th ed. New York, NY: Routledge/Taylor & Francis; 2014.
23. Kuzma JW, Bohnenblust SE. Basic Statistics for the Health Sciences. 5th ed. Boston, MA: McGraw-Hill; 2005.
24. Davis CS. Statistical Methods for the Analysis of Repeated Measurements. New York, NY: Springer; 2002.
25. Daniel WW. Biostatistics: A Foundation for Analysis in the Health Sciences. 9th ed. Hoboken, NJ: John Wiley and Sons; 2009.
26. Kutner MH, Nachtsheim C, Neter J, Li W. Applied Linear Statistical Models. 5th ed. Boston, MA: McGraw-Hill/Irwin; 2005.
27. Shi L. Health Services Research Methods. 2nd ed. Delmar Cengage Learning; 2008.
28. Ghasemi A, Zahediasl S. Normality tests for statistical analysis: a guide for non-statisticians. Int J Endocrinol Metab. 2012;10(2):486–489.
Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.