• Users Online: 134
  • Home
  • Print this page
  • Email this page
Home About us Editorial board Search Ahead of print Current issue Archives Submit article Instructions Subscribe Contacts Login 


 
 Table of Contents  
ORIGINAL ARTICLE
Year : 2019  |  Volume : 16  |  Issue : 1  |  Page : 43-48

Reliability and validity of a disease-specific quality of life tool in Nigerian patients with heart failure


1 Department of Clinical Sciences, Division of Internal Medicine, Limi Hospital Limited, Abuja-FCT, Ibadan, Oyo State, Nigeria
2 Department of Medicine, Division of Cardiology, University College Hospital, Ibadan, Oyo State, Nigeria

Date of Submission13-Apr-2019
Date of Acceptance12-Jun-2019
Date of Web Publication22-Oct-2019

Correspondence Address:
Dr. Iseko Iyoko Iseko
1487, LM Iseko Street, Off Constitution Avenue, Central Area, Abuja-FCT
Nigeria
Login to access the Email id

Source of Support: None, Conflict of Interest: None


DOI: 10.4103/njc.njc_5_19

Rights and Permissions
  Abstract 


Background: Heart failure results in varying limitation in activities of daily living, and consequently on the individual's quality of life. There is a thus a need for a reliable and valid tool to assess HRQoL in Nigerian patients.
Aims and Objectives: To determine the health-related quality of life of Nigerian patients with heart failure using the Minnesota Living with Heart Failure Questionnaire (MLHFQ), its validity and reliability, and its relationship between to disease severity as defined by the NYHA class.
Materials and Methods: 106 consecutive stable participants diagnosed with HF were recruited and studied. The MLHFQ was self-administered by recruited participants and NYHA class was determined by clinical evaluation. Appropriate descriptive and inferential statistical analysis was subsequently carried out.
Results: The mean age (SD) of the subjects was 56.8 (13.4) years. 41.6% of HF subjects had suboptimal HRQoL in varying degrees with a median score of 19. The MLHFQ was found to be reliable exhibiting high internal consistency (Cronbach's alpha reliability coefficient - 0.928), similarly high for each of the subscale domains and there was a significant relationship between the MLHFQ and the NYHA class (rho: 0.7706; P < 0.001), a finding that also validates the MLHFQ.
Conclusion: HRQoL evaluation in Nigerian participants with HF using the MLHFQ should be considered in routine practice as it is a reliable and valid assessment with possible significant clinical utility.

Keywords: Heart failure, Minnesota Living with Heart Failure Questionnaire, New York Heart Association class, quality of life, reliability, validity


How to cite this article:
Iseko II, Adebiyi AA, Falase AO. Reliability and validity of a disease-specific quality of life tool in Nigerian patients with heart failure. Nig J Cardiol 2019;16:43-8

How to cite this URL:
Iseko II, Adebiyi AA, Falase AO. Reliability and validity of a disease-specific quality of life tool in Nigerian patients with heart failure. Nig J Cardiol [serial online] 2019 [cited 2019 Nov 15];16:43-8. Available from: http://www.nigjcardiol.org/text.asp?2019/16/1/43/269652




  Introduction and Background Top


Despite advances in management and new treatment options for heart failure (HF), the mortality remains marked[1] and it has a significant effect on activities of daily living through varying degrees of functional capacity limitation which affect the health-related quality of life (HRQoL), an important index of morbidity. Although reliable estimates are lacking in Nigeria and many other countries, HF is estimated to have a 2%–3% prevalence in the adult population, increasing with age.[2] Health-related quality of life (HRQoL) is a major goal in the context of therapeutic and preventive cardiology,[3] and in recent years, the HRQoL has emerged as a valid measurable endpoint and indeed one of the indices for evaluating efficacy of the treatment of HF. Its emergence is likely not unrelated to increasing longevity of populations and the consequent changing demographics, a trend that has seen general interest in how to achieve “goodness” of life.[4],[5]

The HRQoL can be typically evaluated using either generic or disease-specific instruments.[4] The generic instruments, being less specific to a particular disease, apply to a wide variety of patient groups with a broad spectrum of domains relevant to HRQoL[6] while disease-specific tools focus on areas specific to primary interest as they affect the HRQoL, in this case, HF. Of the disease-specific tools, the Minnesota Living with Heart Failure Questionnaire (MLHFQ), a disease-specific questionnaire developed by Rector et al.[7] in 1987, is considered as the tool most commonly used in clinical trials in which HRQoL is a primary or secondary end point[3] and has documented positive appraisal.[8],[9],[10] Although concerns have been raised about the MLHFQ in terms of the patients' ability to separate symptoms and impairments related to HF from other comorbidities,[11] the measure has been found to be valid in comparison with other health outcome scales.[11] It also has documented validity, reliability, and sensitivity to change (responsiveness) from various studies;[7],[8],[9],[12] however, there is a dearth of studies among Nigerian patients with HF.


  Methodology Top


Over a 12-month period, 106 consecutive stable participants diagnosed with HF (New York Heart Association [NYHA] Class I–III), based on the Framingham criteria[13] (concurrent presence of either two major or one major and two minor criteria), for at least 1 month attending the Cardiology Outpatient Clinic of University College Hospital, Ibadan, were recruited and studied.

Participants with acute HF and acute decompensation of chronic HF or NYHA Class IV were excluded from the study. Similarly, nonconsenting participants and those with a psychiatric diagnosis, for example, schizophrenia and depression were also excluded from the study.

The MLHFQ was self-administered by recruited participants with the help of a trained research assistant where necessary. The degree of help given by the candidate or assistant depended on each patient's ability to read, write, and comprehend the questions. The participants answered the questions without being influenced by others such as their spouse or family members as much as possible. The MLHFQ which contains 21 items with 6-point response scales ranging from 0 to 5 was scored by summating the responses to all 21 questions to produce the total (global) HRQoL score. The possible range of the total HRQoL score is from 0 to 105, with a higher score indicating poorer HRQoL.[7],[12] In addition, three subscale scores (dimensions) which reflect the physical dimension (items 2, 3, 4, 5, 6, 7, 12, and 13), emotional dimension (items 17, 18, 19, 20, and 21), and the other items are related to financial and medication side effects, and lifestyle considerations (overall dimensions) were scored by simple summation to further characterize the effect of HF on a patient's life (43, 70).

Data generated from the study were entered into the Statistical Package for the Social Sciences (SPSS) version 20.0 (IBM Corp., NY, USA). Descriptive statistics were carried out for continuous variables to characterize the sample through means and standard deviations (SDs). HRQoL as measured by MLHFQ was further analyzed to assess the reliability of each of its items and domains by obtaining and examining the Cronbach's alpha reliability coefficient. Validity of the MLHFQ construct, in the absence of a gold standard for HRQoL, was assessed by correlations with NYHA class. The Kruskal–Wallis test was used to assess the MLHFQ scores of the three functional classes, followed by a post hoc Bonferroni test to establish the differences among them, if any.

Being a nonbiological variable, the HRQoL did not follow a normal distribution, and this was demonstrated by the Shapiro–Wilk test. Consequent on this, median scores which represent a better measure of central tendency in this case were predominantly used as opposed to mean scores. Behlouli et al.'s[14] MLHFQ score cutoffs determined by neural network approach were adopted where a score of <24 on the MLHFQ represents a good HRQoL, a score between 24 and 45 represents a moderate HRQoL, and a score >45 represents a poor HRQoL.


  Results Top


The mean age (SD) of the participants was 56.8 (13.4) years with age ranging between 21 and 85 years and the frequency of males with HF being about twice that of females.

About half of the participants had postprimary school education (secondary, tertiary, and posttertiary), and most participants finance their health-care out-of-pocket expenditure [See [Table 1]].
Table 1: Baseline sociodemographic characteristics of participants

Click here to view


Health-related quality of life in participants with heart failure

[Table 2] shows the distribution of HRQoL and its domains. Using Behlouli et al.'s[14] MLHFQ classification, 62 (58.5%, 95% confidence interval (CI): 47.0–69.0) participants were classified as having good quality of life, while 22 (20.8%, 95% CI: 11.9–30.0) of the participants had moderate and 22 (20.8%, 95% CI: 11.9–30.0) had poor quality of life irrespective of HF etiology.
Table 2: Global and subscale quality of life dimensions in participants with heart failure

Click here to view



  Reliability of the Minnesota Living With Heart Failure Quality of Life Questionnaire Top


The Cronbach's alpha reliability coefficient, which is used to estimate the reliability of psychometric tests[15],[16] such as the MLHFQ, was used to assess reliability by determining the internal consistency. It determines how closely related a set of items are as a group and to know that the instrument used will always elicit a consistent and reliable response even if questions were replaced with other similar questions. It is a function of the number of items in the test, the average covariance between item pairs, and the variance of the total score.

The value of Cronbach's alpha varies from 0 to 1, with higher values being more desirable. A reliability of 0.70 or higher is generally regarded as an acceptable reliability coefficient.[16]

The three domain scores of MLHFQ had an overall reliability coefficient (Cronbach's alpha) of 0.928 indicating high degree of internal consistency. The largest domain – global correlation coefficients – was noticed for physical domain (rho: 0.931), followed by overall domain (rho: 0.870) and emotional domain (rho: 0.801).

Reliability coefficient of individual domains showed that physical domain had the highest internal consistency (Cronbach's alpha: 0.927), followed by emotional domain (Cronbach's alpha: 0.814), while the overall domain had the least internal consistency (Cronbach's alpha: 0.767).

Validity of the Minnesota Living with Heart Failure Quality of Life Questionnaire

Global scores showed a significant difference between different NYHA classes with post hoc testing showing significant between all groups. Participants with different NYHA severity also differed in all domains of the MLHFQ.

NYHA class differences resulted in distinctive domain and global score profiles, as shown in [Figure 1]. These were significant after performing a MANOVA for domain scores across groups (Pillai T2 = 10.165, P < 0.001).
Figure 1: Profile of the Minnesota Living with Heart Failure Questionnaire scores for the New York Heart Association classes

Click here to view


[Table 3] shows a statistically significant relationship between the NYHA class and the MLHFQ score (global score and subscale domain scores). The post hoc Bonferroni test done showed that the differences observed between the individual functional classes were similarly significant (i.e., between Classes I and II and between Classes II and III).
Table 3: The Minnesota Living with Heart Failure Questionnaire scores across various NYHA classes (Kruskal–Wallis test)

Click here to view



  Discussion Top


Growing in its relevance and application in chronic diseases, the HRQoL in HF plays an important role as a treatment endpoint for intervention studies and also as a prognostic tool in the routine management of HF participants.[3] Compared to other chronic diseases, the HRQoL is affected relatively more in HF[17] and is an important patient-reported goal of treatment compared to the duration of life.[18],[19]

The HRQoL in participants as assessed by the MLHFQ with HF in this study was poor in 20.8% of participants while an additional 20.8% had moderate HRQoL, making a total of 41.6% of participants with suboptimal HRQoL in varying degrees. Quality of life has been reported to be affected significantly to varying degrees in participants with HF from other studies, and in some, it has also been shown to have prognostic value.[20],[21],[22],[23],[24] In a study from Lagos, Mbakwem[22] reported 27.36% of HF participants with suboptimal quality of life using the Kansas City Cardiomyopathy Questionnaire (KCCQ), a disease-specific questionnaire.

However, with a group mean of 23.8 and median of 19.0, the global HRQoL was relatively low, a finding which was similar to research from Spain which reported a median of 28.[10] This may perhaps be the consequence of nondisclosure of sexual-related features and relative amelioration or alteration in perspective on emotional issues following religious and spiritual inclinations of the populace. This may possibly suggest that the questionnaire might need some adaptation to better represent the prevailing needs and values to the Nigerian context.

All the measured domains of quality of life in HF participants are affected with the physical domain (median: 8; interquartile range: 3–17) being affected more than emotional domain (median: 3; interquartile range: 0–3). While it is sometimes suggested that the emotional domain of quality of life may be less affected in Nigerians because of the extended family support systems that exist,[25],[26] no statistically significant correlation or association was found between living alone or the number of caregivers and the emotional domain of the HRQoL in this study.

Literature documenting appraisal of the MLHFQ tool in Nigerian participants with HF is at a dearth, and this study documents the reliability using statistical tools and the validity by determining the relationship of the MLHFQ to existing traditional clinical outcome variables in routine use in Nigeria such as the NYHA classification and the distance covered during a standard 6-min walk test.

Reliability of the Minnesota Living with Heart Failure Questionnaire

The MLHFQ reliability coefficient was measured by examining the Cronbach's alpha, a reliability estimate for psychometric tests,[27] for the global MLHFQ score and all its subscale domains. In this study, the global MLHFQ score had a Cronbach's alpha (reliability coefficient) of 0.928 indicating high internal consistency, and it was similarly high for each of the subscale domains (0.801–0.931). Having demonstrated a high internal consistency coefficient, the inter-item correlation of the MLHFQ indicated the absence of repetitive questions, i.e., questions on the MLHFQ that essentially tests the same thing. Such high internal consistency as seen in this study suggests that the total MLHFQ score measures a single construct, in this case presumed to be the effect of HF on a patient's quality of life. This finding is a similar finding of the reliability coefficient reported by several other authors in the appraisal of the MLHFQ[8],[9],[28],[29] and the item-total correlation coefficients in the 21 items, and discriminative ability in all items and subscale were within acceptable limits as similarly reported by Heo et al.[29] in a study examining the psychometric properties of the MLHFQ.


  Validity of the Minnesota Living With Heart Failure Questionnaire and Its Relationship to the New York Heart Association Class Top


The validity of the MLHFQ as with other HRQoL tools refers to the assessment of whether the tool is measuring what it is supposed to measure.[6] The absence of a gold standard for HRQoL means that the validity determination of the HRQoL tool necessitates the examination of the MLHFQ discriminant validity by determining whether the dimensions of the MLHFQ were able to distinguish between participants with different levels of HF severity as determined by other variables (in this study, NYHA).[6],[30] The correlation between the MLHFQ score (global and subdomains) and the NYHA class in this study demonstrated statistical significance (P < 0.001) more so with the physical domain score as expected. This relationship is rather unsurprising as the NYHA classification is essentially a provider-derived classification scheme to categorize participants in terms of symptoms associated with daily activities. The trend toward higher MLHFQ scores with worsening NYHA functional class in participants with HF was demonstrated [Figure 2], and this finding was statistically significant. Such a trend is a characteristic finding in the MLHFQ validation studies and was demonstrated similarly by Parajón et al.[10] and Morcillo et al.[31] in studies from Spain.
Figure 2: Box plot showing the Minnesota Living with Heart Failure Questionnaire scores for each New York Heart Association functional class

Click here to view


The relationship of NYHA class to HRQoL was not clearly demonstrated by Ola et al.[24] although their study was carried out with a generic HRQoL questionnaire, and this may account for that finding. However, while using another disease-specific HRQoL measure (KCCQ), a statistically significant correlation between NYHA class and HRQoL was demonstrated by Mbakwem.[22]

Although at the time of this study there has been no available published MLHFQ validity data from Nigeria, the validity of this tool has been demonstrated in previous studies carried out in other parts of the world.[7],[8],[10],[28],[32]


  Conclusion Top


Being a patient-reported outcome measure, HRQoL assessment in HF corroborates with the global trend towards patient-centered care as a goal of healthcare systems,[33],[34] and its addition to the traditional approach to patient evaluation is therefore of increasing importance and value.

The evaluation of HRQoL in Nigerian participants with HF using the MLHFQ is a reliable and valid assessment with significant clinical utility, and thus, it should be considered in routine practice and research evaluations of participants with HF.

Informed consent

Informed consent was obtained from all individual participants included in the study.

Declaration of patient consent

The authors certify that they have obtained all appropriate patient consent forms. In the form, the patients have given their consent for their images and other clinical information to be reported in the journal. The patients understand that their names and initials will not be published and due efforts will be made to conceal their identity, but anonymity cannot be guaranteed.

Financial support and sponsorship

This was a self-funded study.

Conflicts of interest

There are no conflicts of interest.



 
  References Top

1.
Gustafsson F, Arnold JM. Heart failure clinics and outpatient management: Review of the evidence and call for quality assurance. Eur Heart J 2004;25:1596-604.  Back to cited text no. 1
    
2.
López-Sendón J. The heart failure epidemic. Medicographia 2011;33:3639.  Back to cited text no. 2
    
3.
Coelho R, Ramos S, Prata J, Bettencourt P, Ferreira A, Cerqueira-Gomes M. Heart failure and health related quality of life. Clin Pract Epidemiol Ment Health 2005;1:19.  Back to cited text no. 3
    
4.
Bowling A. Health-Related Quality of Life: Conceptual Meaning, Use and Measurement. Measuring Disease: A review of Disease-Specific Quality of Life Measuring Scales. 2nd ed. Buckingham: Open University Press; 2001. p. 21.  Back to cited text no. 4
    
5.
Berlin RB Jr., Schatz BR. Monitoring of quality of life in congestive heart failure populations. Congest Heart Fail 2001;7:13-21.  Back to cited text no. 5
    
6.
Guyatt GH. Measurement of health-related quality of life in heart failure. J Am Coll Cardiol 1993;22:185A-91A.  Back to cited text no. 6
    
7.
Rector T, Kubo S, Cohn J. Patients' self-assessment of their congestive heart failure, II: Content, reliability, and validity of a new measure, the Minnesota living with heart failure questionnaire. Heart Fail 1987;3:198-209.  Back to cited text no. 7
    
8.
Rector TS, Cohn JN. Assessment of patient outcome with the Minnesota living with heart failure questionnaire: Reliability and validity during a randomized, double-blind, placebo-controlled trial of pimobendan. Pimobendan multicenter research group. Am Heart J 1992;124:1017-25.  Back to cited text no. 8
    
9.
Rector TS, Kubo SH, Cohn JN. Validity of the Minnesota living with heart failure questionnaire as a measure of therapeutic response to enalapril or placebo. Am J Cardiol 1993;71:1106-7.  Back to cited text no. 9
    
10.
Parajón T, Lupón J, González B, Urrutia A, Altimir S, Coll R, et al. Use of the Minnesota living with heart failure quality of life questionnaire in Spain. Rev Esp Cardiol 2004;57:155-60.  Back to cited text no. 10
    
11.
Dunderdale K, Thompson DR, Miles JN, Beer SF, Furze G. Quality-of-life measurement in chronic heart failure: Do we take account of the patient perspective? Eur J Heart Fail 2005;7:572-82.  Back to cited text no. 11
    
12.
Rector T, Francis G, Cohn J. Patients' self-assessment of their congestive heart failure I: Patient perceived dysfunction and its poor correlation with maximal exercise tests. Heart Fail 1987;3:192-6.  Back to cited text no. 12
    
13.
McKee PA, Castelli WP, McNamara PM, Kannel WB. The natural history of congestive heart failure: The Framingham study. N Engl J Med 1971;285:1441-6.  Back to cited text no. 13
    
14.
Behlouli H, Feldman DE, Ducharme A, Frenette M, Giannetti N, Grondin F, et al. Identifying relative cut-off scores with neural networks for interpretation of the Minnesota living with heart failure questionnaire. Conf Proc IEEE Eng Med Biol Soc 2009;2009:6242-6.  Back to cited text no. 14
    
15.
Santos JR. Cronbach's alpha: A tool for assessing the reliability of scales. J Ext 1999;37:1-5.  Back to cited text no. 15
    
16.
Nunnally J. Psychometric Theory. New York: McGraw-Hill; 1978.  Back to cited text no. 16
    
17.
Juenger J, Schellberg D, Kraemer S, Haunstetter A, Zugck C, Herzog W, et al. Health related quality of life in patients with congestive heart failure: Comparison with other chronic diseases and relation to functional variables. Heart 2002;87:235-41.  Back to cited text no. 17
    
18.
Lewis EF, Johnson PA, Johnson W, Collins C, Griffin L, Stevenson LW. Preferences for quality of life or survival expressed by patients with heart failure. J Heart Lung Transplant 2001;20:1016-24.  Back to cited text no. 18
    
19.
Stanek EJ, Oates MB, McGhan WF, Denofrio D, Loh E. Preferences for treatment outcomes in patients with heart failure: Symptoms versus survival. J Card Fail 2000;6:225-32.  Back to cited text no. 19
    
20.
Hoekstra T, Lesman-Leegte I, van Veldhuisen DJ, Sanderman R, Jaarsma T. Quality of life is impaired similarly in heart failure patients with preserved and reduced ejection fraction. Eur J Heart Fail 2011;13:1013-8.  Back to cited text no. 20
    
21.
Hobbs FD, Kenkre JE, Roalfe AK, Davis RC, Hare R, Davies MK, et al. Impact of heart failure and left ventricular systolic dysfunction on quality of life: A cross-sectional study comparing common chronic cardiac and medical disorders and a representative adult population. Eur Heart J 2002;23:1867-76.  Back to cited text no. 21
    
22.
Mbakwem AC. Comparative analysis of the quality of life of heart failure patients in South Western Nigeria. World J Cardiovasc Dis 2013;3:146-53.  Back to cited text no. 22
    
23.
Rogers WJ, Johnstone DE, Yusuf S, Weiner DH, Gallagher P, Bittner VA, et al. Quality of life among 5,025 patients with left ventricular dysfunction randomized between placebo and enalapril: The studies of left ventricular dysfunction. The SOLVD investigators. J Am Coll Cardiol 1994;23:393-400.  Back to cited text no. 23
    
24.
Ola BA, Adewuya AO, Ajayi OE, Akintomide AO, Oginni OO, Ologun YA. Relationship between depression and quality of life in Nigerian outpatients with heart failure. J Psychosom Res 2006;61:797-800.  Back to cited text no. 24
    
25.
El-Islam EF. A better outlook for schizophrenics living in extended families. Br J Psychiatry 1979;135:343-7.  Back to cited text no. 25
    
26.
Abiola DI, Morakinyo O, Ibrahim A. Quality of life of patients with schizophrenia in North-Western Nigeria. Internet J Psychiatry 2012;2. Available from: http://ispub.com/IJPSY/2/1/2975. [Last accessed on 2015 Jan 27].  Back to cited text no. 26
    
27.
Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika 1951;16:297-334.  Back to cited text no. 27
    
28.
Riegel B, Moser DK, Glaser D, Carlson B, Deaton C, Armola R, et al. The Minnesota living with heart failure questionnaire: Sensitivity to differences and responsiveness to intervention intensity in a clinical population. Nurs Res 2002;51:209-18.  Back to cited text no. 28
    
29.
Heo S, Moser DK, Riegel B, Hall LA, Christman N. Testing the psychometric properties of the Minnesota living with heart failure questionnaire. Nurs Res 2005;54:265-72.  Back to cited text no. 29
    
30.
Naveiro-Rilo JC, Diez-Juárez DM, Romero Blanco A, Rebollo-Gutiérrez F, Rodríguez-Martínez A, Rodríguez-García MA. Validation of the Minnesota living with heart failure questionnaire in primary care. Rev Esp Cardiol 2010;63:1419-27.  Back to cited text no. 30
    
31.
Morcillo C, Aguado O, Delás J, Rosell F. Utility of the Minnesota living with heart failure questionnaire for assessing quality of life in heart failure patients. Rev Esp Cardiol 2007;60:1093-6.  Back to cited text no. 31
    
32.
Santos JJ, Plewka JE, Brofman PR. Quality of life and clinical indicators in heart failure: A multivariate analysis. Arq Bras Cardiol 2009;93:159-66.  Back to cited text no. 32
    
33.
Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, D.C.: National Academy Press; 2001. p. 36. Available from: http://www.nap.edu/catalog.php?record_id=10027. [Last accessed on 2015 Jan 27].  Back to cited text no. 33
    
34.
Yong PL, Olsen L, McGinnis JM, editors. Institute of Medicine. Value in Health Care: Accounting for Cost, Quality, Safety, Outcomes, and Innovation. Washington, DC: National Academies Press (US); 2010. Available from: http://www.ncbi.nlm.nih.gov/books/NBK50929/. [Last accessed on 2015 Jan 27].  Back to cited text no. 34
    


    Figures

  [Figure 1], [Figure 2]
 
 
    Tables

  [Table 1], [Table 2], [Table 3]



 

Top
 
 
  Search
 
Similar in PUBMED
   Search Pubmed for
   Search in Google Scholar for
 Related articles
Access Statistics
Email Alert *
Add to My List *
* Registration required (free)

 
  In this article
Abstract
Introduction and...
Methodology
Results
Reliability of t...
Discussion
Validity of the ...
Conclusion
References
Article Figures
Article Tables

 Article Access Statistics
    Viewed57    
    Printed2    
    Emailed0    
    PDF Downloaded16    
    Comments [Add]    

Recommend this journal


[TAG2]
[TAG3]
[TAG4]