ࡱ > b d Y Z [ \ ] ^ _ ` a [ X bjbj A pa!\pa!\4& , 0 0 d % | ) f9 |9 |9 |9 G L +N z z z z z z z $ e F z 3O D \ G 3O 3O z |9 |9 ? | S S S 3O p |9 |9 rl > S 3O z S S |b l e |9 Կ^ Q c t ^l | 0 | \d R d T e e *e 4 3O 3O S 3O 3O 3O 3O 3O z z S 3O 3O 3O | 3O 3O 3O 3O 3O 3O 3O 3O 3O 3O 3O 3O 3O 0 > n : A systematic review of psychological correlates of HIV testing intention
Michael Evangeli, Krissie Ferris, Natalie M. Kenney, Laura L. E. Baker, Bethanie Jones, and Abigail L. Wroe1
1 Department of Psychology, Royal Holloway University of London, Egham, Surrey, UK
Correspondence to Dr. Michael Evangeli, Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK. Tel: 00441784443851. Fax: 01784472746. Email: HYPERLINK "mailto:michael.evangeli@rhul.ac.uk" michael.evangeli@rhul.ac.uk
Abstract
Undiagnosed HIV infection is associated with onward HIV transmission and delays in accessing HIV care and treatment. As a significant proportion of HIV tests are self-initiated, it is important to assess correlates of the intention to test for HIV. Psychological correlates of HIV testing intention are more likely to be the feasible target of interventions than structural determinants. A systematic review of psychological correlates of HIV testing intention was conducted. Twenty studies were included in the review, covering a range of populations and geographical regions. The most commonly assessed variables were HIV risk perception and HIV knowledge rather than HIV test-specific psychological factors. There was evidence that HIV risk perception and pro-testing attitudes were consistently associated with HIV testing intention across a number of studies. There is a need for longitudinal designs, including experimental studies, allowing for more confident casual inferences to be made. Theoretical, research and practice implications are outlined.
Key words: HIV testing; intention; willingness; systematic review; psychological
Introduction
For people who are HIV positive, accurate diagnosis via HIV testing is a prerequisite for receiving treatment and care. Early diagnosis and access to treatment is associated with a reduced likelihood of onward transmission ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_23" \o "Fox, 2009 #121" Fox et al., 2009), better response to antiretroviral treatment (ART), and reduced mortality and morbidity ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_15" \o "Egger, 2002 #128" Egger et al., 2002). However, many people living with HIV are unaware of their status, with the World Health Organisation (WHO) estimating that less than half of those infected with HIV have been diagnosed ADDIN EN.CITE WHO2014208(WHO, 2014)20820827WHOGlobal update of the health sector response to HIV, 20142014GenevaWorld Health Organisation( HYPERLINK \l "_ENREF_60" \o "WHO, 2014 #208" WHO, 2014). The growing availability of ART, and the increasing evidence of its individual and public health benefits, reinforces the need to scale up testing ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_24" \o "Granich, 2009 #1164" Granich, Gilks, Dye, De Cock, & Williams, 2009), and to research the factors influencing testing decisions ADDIN EN.CITE WHO/UNAIDS2012129(WHO/UNAIDS, 2012)12912927WHO/UNAIDSService delivery approaches to HIV testing and counselling (HTC): A strategic programme framework. 2012GenevaWorld Health Organisation( HYPERLINK \l "_ENREF_63" \o "WHO/UNAIDS, 2012 #129" WHO/UNAIDS, 2012).
Current WHO recommendations state that HIV testing should be informed, voluntary and confidential ADDIN EN.CITE WHO2015953(WHO, 2015)95395327WHOConsolidated guidelines on HIV testing services2015GenevaWorld Health Organisation( HYPERLINK \l "_ENREF_61" \o "WHO, 2015 #953" WHO, 2015). Historically, voluntary counselling and testing (VCT) has been the dominant model, with individuals actively seeking HIV tests from healthcare or community facilities ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_57" \o "van Rooyen, 2013 #131" van Rooyen et al., 2013). Despite initiatives to offer routine testing in some settings ADDIN EN.CITE WHO/UNAIDS2007130(WHO/UNAIDS, 2007)13013027WHO/UNAIDSGuidance on provider-initiated HIV testing and counselling in health facilities. 2007GenevaWorld Health Organisation( HYPERLINK \l "_ENREF_62" \o "WHO/UNAIDS, 2007 #130" WHO/UNAIDS, 2007), there remain a significant proportion of the HIV positive population whose diagnosis still relies on VCT. Recent self-testing initiatives have further highlighted the importance of individuals HIV testing decision-making ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_33" \o "Johnson, 2014 #39" Johnson et al., 2014).
Social cognition models, including the Theory of Planned Behavior (TPB) ADDIN EN.CITE Ajzen1985146(Ajzen, 1985)1461465Ajzen, I.Kuhl, J.,Beckmann, J.From intentions to action: A thoery of planned behavior.Action-control: From cognition to behavior11-391985HeidelbergSpringer( HYPERLINK \l "_ENREF_2" \o "Ajzen, 1985 #146" Ajzen, 1985) and the Health Belief Model (HBM) ADDIN EN.CITE Rosensto.Im197461(Rosensto.Im, 1974)616117Rosensto.ImUniv Michigan,Sch Publ Hlth,Ann Arbor,Mi 48104Historical Origins of Health Belief ModelHealth Education MonographsHealth Educ Quart328-3352419740073-1455WOS:A1974V233700002<Go to ISI>://WOS:A1974V233700002English( HYPERLINK \l "_ENREF_50" \o "Rosensto.Im, 1974 #61" Rosensto.Im, 1974) have highlighted the importance of individual proximal determinants of health behaviour. Many models, including the TPB, suggest that the likelihood of performing a behaviour depends on the strength of intention, which in turn is influenced by other psychological factors (e.g., behavioural attitudes). Intention may be a particularly important predictor of HIV testing in situations when testing is under voluntary control. In addition, intention-driven processing has been argued to be relevant for important decisions and behaviours in novel situations ADDIN EN.CITE Ajzen2011994(Ajzen, 2011)99499417Ajzen, I.The theory of planned behaviour: reactions and reflectionsPsychol HealthPsychology & healthPsychol HealthPsychology & healthPsychol HealthPsychology & health1113-27269*BehaviorCharacter*Decision MakingEmotionsHabitsHumansIndividuality*IntentionInternal-External Control*MotivationProblem Solving*Psychological TheorySocial Conformity2011Sep1476-8321 (Electronic)
0887-0446 (Linking)21929476http://www.ncbi.nlm.nih.gov/pubmed/2192947610.1080/08870446.2011.613995( HYPERLINK \l "_ENREF_3" \o "Ajzen, 2011 #994" Ajzen, 2011) (p1122), characteristic of many HIV testing situations. Indeed, there is evidence that HIV testing intention is related to future HIV testing ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_45" \o "Mirkuzie, 2011 #646" Mirkuzie, Sisay, Moland, & Astrom, 2011). Whilst examining correlates of HIV testing is important, conclusions are limited by cross sectional designs where predictors are measured using current time-frames and outcomes refers to a retrospective event (lifetime testing) ADDIN EN.CITE Evangeli2016998(Evangeli, Pady, & Wroe, 2016)99899817Evangeli, M.Pady, K.Wroe, A. L.Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK. michael.evangeli@rhul.ac.uk.
Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK.Which Psychological Factors are Related to HIV Testing? A Quantitative Systematic Review of Global StudiesAIDS BehavAIDS and behaviorAIDS BehavAIDS and behaviorAIDS BehavAIDS and behavior880-9182042016Apr1573-3254 (Electronic)
1090-7165 (Linking)26566783http://www.ncbi.nlm.nih.gov/pubmed/26566783479926710.1007/s10461-015-1246-0( HYPERLINK \l "_ENREF_18" \o "Evangeli, 2016 #998" Evangeli, Pady, & Wroe, 2016). This makes establishing the likely direction of any causal relationships difficult. Testing intention refers to a future anticipated event that can be investigated in relation to current predictors (e.g., HIV knowledge) and future actual events (e.g., testing) with a greater degree of confidence about causality.
We examined the relationship between psychological factors and HIV testing intention. While researching demographic and structural associations with testing and testing intention is necessary for targeting interventions to appropriate populations ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_5" \o "Andrews, 2011 #148" Andrews, 2011; HYPERLINK \l "_ENREF_32" \o "Jin, 2002 #164" Jin et al., 2002), it is crucial also to understand psychological correlates, as these factors are potentially modifiable.
Methods
Study eligibility criteria
We followed PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses) Statement guidelines [1]. Studies were included if they:
Used quantitative designs;
Included participants with the capacity to decide to test for HIV. Studies of populations requiring parental/guardian consent for HIV testing (e.g. children or individuals with moderate to severe learning disabilities), or for whom HIV testing was mandatory (e.g. some state prisoners in the U.S.A.) were excluded;
Assessed psychological variables, relating to an individuals internal state (e.g. feelings or beliefs);
Included the variable of intention or willingness to test.
Sources of information
Studies published in peer-reviewed journals were retrieved from Pubmed/Medline, PsycINFO, Web of Science and Cochrane Library. Papers were also searched from conference proceedings (International AIDS conference, AIDS Impact, International AIDS Society Conference), and the Networked Digital Library of Theses and Dissertations (NDLTD). We restricted searches to studies conducted since January 1, 1996. This was due to biomedical advances in 1996, leading to the uptake of effective antiretroviral regimens for HIV/AIDS.
Search strategy
Searches were conducted using combinations of the following terms: HIV testing intention, HIV testing readiness, HIV testing willingness, HIV testing preparedness, HIV testing motivation, psychological, psychosocial, psychiatric, cognitive, affective, behavioural (behavioral), psychopathology, mood, beliefs, illness perception and illness representation. HIV testing, HIV testing readiness, HIV testing willingness and HIV testing preparedness, were searched as keywords in titles; the psychological terms were also searched for as keywords in abstracts.
Data collection
Following PRISMA guidelines ADDIN EN.CITE Moher200984(Moher, Liberati, Tetzlaff, Altman, & Group, 2009)848417Moher, D.Liberati, A.Tetzlaff, J.Altman, D. G.Prisma GroupOttawa Methods Centre, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada. dmoher@ohri.caPreferred reporting items for systematic reviews and meta-analyses: the PRISMA statementPLoS MedPLoS medicinePLoS MedPLoS medicinePLoS MedPLoS medicinee100009767Evidence-Based Practice/standardsHumans*Meta-Analysis as TopicPeriodicals as Topic/standardsPublication BiasPublishing/*standardsQuality Control*Review Literature as Topic*Terminology as Topic2009Jul 211549-1676 (Electronic)
1549-1277 (Linking)19621072http://www.ncbi.nlm.nih.gov/pubmed/19621072270759910.1371/journal.pmed.1000097( HYPERLINK \l "_ENREF_46" \o "Moher, 2009 #84" Moher, Liberati, Tetzlaff, Altman, & Group, 2009), data collection process had four stages (see Figure 1). One reviewer (xx or xx) carried out searches for the identification of studies, using pre-specified search criteria. All duplications were removed. Searches were completed on December 7th 2014. Two reviewers (xx or xx and xx or xx) independently screened remaining titles and abstracts for eligibility. Articles considered relevant by either reviewer were retrieved in full text. The two reviewers then independently assessed eligibility of the retrieved articles. Any disagreements were resolved by a third reviewer (xx or xx).
Figure 1 here
Quality Assessment
The quality of the studies was evaluated. Papers were assessed on two dimensions of external validity and three dimensions of internal validity (see Table I).
Table I here
xx/xx and xx/xx assessed all papers independently before comparing ratings and coming to an agreement where this was initially lacking. Ongoing disagreements were resolved by xx or xx.
Results
Twenty studies were included (see Table II).
Table II here
Study Characteristics and participants
All studies were published between 2005 and 2014, six in the USA, six in Africa, two in Europe, five in Asia, and one in South America. All but one used a cross-sectional design. A total of 18,323 participants were included across studies, with sample sizes from 92 to 4002 (median 436.5; IQR 258.5 1119.5). A wide range of populations were sampled. Four studies sampled migrants or ethnic minorities, four sampled students HYPERLINK \l "_ENREF_19" \o "Diteweg, 2013 #1072" , four used occupational populations HYPERLINK \l "_ENREF_26" \o "Kakoko, 2006 #1034" , two sampled men who have sex with men. The remainder used a range of other populations HYPERLINK \l "_ENREF_28" \o "Smith, 2006 #1059" HYPERLINK \l "_ENREF_35" \o "Yuan, 2012 #1070" .
HIV Testing Intention Measurement
Only six studies specified a timescale for testing intention (e.g., intending to test in the next 3 months). Ten studies used Likert scale responses, of which three used more than one item.
Relationship between Psychological Variables and HIV Testing Intention
Several studies using health behaviour theories to select variables, commonly the Theory of Planned Behavior and Health Belief Model. Findings are discussed for individual variables where these appeared in two or more studies. The majority of studies used multi-item scales. A summary of findings is presented in Table III.
Table III here
HIV Testing-Related Psychological Variables
Perceived Benefits of Testing/Pro-testing Attitudes
Nine studies measured the perceived benefits of testing or pro-testing attitudes, two of which measured both positive and negative attitudes combined. Six of these studies found positive relationships between pro-testing attitudes and testing intention across a range of risk groups. One study found a positive relationship between behavioural beliefs about testing and intention, but not between testing attitudes and intention. One study found a positive relationship between some pro-testing attitudes and not others. Only one study failed to show any positive relationship(s) between pro-testing attitudes and intention.
Perceived Barriers to Testing/Cons of Testing
Six studies measured perceived barriers to testing or cons of testing (none in Africa). Regardless of population, there was little evidence of relationships between perceived barriers to testing and intention. Only two of the six studies found a relationship between any barriers and intention and within these two studies there were also non-significant relationships between specific barriers and intention.
Perceived Behavioural Control/Self-efficacy
Six studies measured perceived behavioural control or self-efficacy, with weak evidence of a relationship with intention. Two studies reported relationships between higher self-efficacy and testing intention and two studies failed to find a relationship. One study reported both an absence of a relationship and an association between lower self-efficacy and intention depending on ethnicity, and one study reported a relationship between high levels of perceived behavioural control and testing intention in univariate but not multivariate analysis. Those studies where a positive relationship were found took place in Africa.
Perceived Norms of Testing
Six studies assessed HIV testing norms, one of which measured both subjective social norms (perceived social pressure to test) and descriptive norms (perceived frequency of testing within a population). Of the five studies assessing subjective social norms, two studies reported significant relationships between normative beliefs and testing intention and three studies failed to find a relationship. Both studies that measured descriptive norms found a positive relationship between with testing intention, although one of these studies also found non-significant relationships between different aspects of descriptive norms and intention. The nature of the study population was not related to the pattern of findings.
Perceived Accessibility and Knowledge of Testing Site
Three studies measured the perceived accessibility of HIV testing and knowledge of HIV testing sites, in two studies using a single item. Significant relationships with testing intention were reported in two of the studies, with no relationship in the third.
Fear of Testing
Four studies measured fear associated with HIV testing with specific items or as part of a broader scale about emotions associated with testing. Only one of the studies showed a relationship between fear of testing (greater fear) and testing intention.
Non Testing HIV-Related Psychological Variables
HIV Knowledge
Eleven studies measured HIV knowledge. Five studies failed to show any relationship between with testing intention. Three studies showed relationships between knowledge and intention for some questions and not others, for one subgroup but not another, or in univariable but not multivariable analysis. Three studies showed relationships between higher levels of HIV knowledge and testing intention. The nature of the study population was not related to the pattern of findings.
Perceived Risk of HIV/susceptibility to HIV
Sixteen studies measured perceived risk of HIV or susceptibility to HIV, usually with single item scales. Eleven studies measured the perceived risk of acquiring HIV in the future/perceived susceptibility. Eight of these studies found a positive relationship between perceived risk and testing intention with three studies failing to find a relationship. In the four studies where it was either unclear whether the measure referred to current or future HIV risk, or both aspects were assessed, all showed a relationship between higher levels of perceived HIV risk and intention. Only one study clearly measured current perceived risk, failing to find a relationship with testing intention. Positive relationships between perceived HIV risk and testing intention were particularly consistent in samples from low HIV prevalence populations.
Peer HIV risk perception
Three studies measured the perceived HIV risk of peers or partners, all failing to find a relationship with testing intention
HIV Stigma
Six studies measured stigma towards people living with HIV (e.g., individual or perceived community prejudiced attitudes). Two studies showed a relationship between lower levels of HIV stigma and testing intention. Three studies failed to find a relationship. One study showed a relationship between some, but not other, aspects of HIV stigma and testing intention. Significant relationships between HIV stigma and testing intention were more likely in samples recruited from higher HIV prevalence populations.
Perceived Severity of HIV
Six studies measured HIV severity. A significant relationship between HIV severity (greater) and testing intention was only seen in two studies, with no effect in four studies. Significant relationships between perceived severity of HIV and intention were more likely in samples recruited from lower HIV prevalence populations.
Non HIV-related variables
Well being
Two studies measured aspects of well-being. Both studies found that greater happiness/life satisfaction were related to testing intention.
Acculturation/integration
Two studies measured acculturation or integration of migrants into the host country. One study found that greater integration was related to testing intention. The other study did not find a relationship between acculturation and testing intention.
Methodological Quality
The methodological quality of studies is summarised in Table IV.
Table IV here
A tick (() signifies that the criterion was met. A cross (x) indicates that the criterion was either not met or it was unclear if the criterion was met. Five studies met both external validity criteria HYPERLINK \l "_ENREF_17" \o "Fernandez, 2005 #1013" and five studies met all internal validity criteria HYPERLINK \l "_ENREF_22" \o "Fako, 2006 #1009" .
Discussion
We carried out a systematic review of psychological correlates of HIV testing intention. Studies covered a wide range of regions and populations. Sample sizes were large, minimising the likelihood that non-significant results were due to a lack of statistical power. Theoretical models guided the selection of variables examined in some studies ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_3" \o "Ajzen, 2011 #994" Ajzen, 2011; HYPERLINK \l "_ENREF_50" \o "Rosensto.Im, 1974 #61" Rosensto.Im, 1974).
Across populations, consistent with a recent review of correlates of HIV testing ADDIN EN.CITE Evangeli2016998(Evangeli et al., 2016)99899817Evangeli, M.Pady, K.Wroe, A. L.Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK. michael.evangeli@rhul.ac.uk.
Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK.Which Psychological Factors are Related to HIV Testing? A Quantitative Systematic Review of Global StudiesAIDS BehavAIDS and behaviorAIDS BehavAIDS and behaviorAIDS BehavAIDS and behavior880-9182042016Apr1573-3254 (Electronic)
1090-7165 (Linking)26566783http://www.ncbi.nlm.nih.gov/pubmed/26566783479926710.1007/s10461-015-1246-0( HYPERLINK \l "_ENREF_18" \o "Evangeli, 2016 #998" Evangeli et al., 2016), there was strong evidence that the perceived benefits of testing/pro-testing attitudes were associated with testing intention, and limited evidence that perceived barriers to testing/cons of testing were associated with testing intention. There was also limited evidence of a relationship between testing self-efficacy/perceived behavioural control and testing intention (although those studies were associations were found all took place in Africa). This is in contrast to evidence of positive relationships between self-efficacy and testing itself ADDIN EN.CITE Evangeli2016998(Evangeli et al., 2016)99899817Evangeli, M.Pady, K.Wroe, A. L.Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK. michael.evangeli@rhul.ac.uk.
Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK.Which Psychological Factors are Related to HIV Testing? A Quantitative Systematic Review of Global StudiesAIDS BehavAIDS and behaviorAIDS BehavAIDS and behaviorAIDS BehavAIDS and behavior880-9182042016Apr1573-3254 (Electronic)
1090-7165 (Linking)26566783http://www.ncbi.nlm.nih.gov/pubmed/26566783479926710.1007/s10461-015-1246-0( HYPERLINK \l "_ENREF_18" \o "Evangeli, 2016 #998" Evangeli et al., 2016). The majority of studies examining the latter association have measured lifetime testing, whereas in the current review, the outcome variable (testing intention) refers to the future. It may be that higher levels of HIV testing self-efficacy are not primarily a determinant of testing and testing intention but a consequence of previous HIV testing. Alternatively, the nature of the population may influence whether self-efficacy affects testing intention.
There was little evidence of a relationship between subjective social norms and testing intention (but stronger evidence for descriptive norms), despite this variable appearing in the TPB model ADDIN EN.CITE Ajzen2011994(Ajzen, 2011)99499417Ajzen, I.The theory of planned behaviour: reactions and reflectionsPsychol HealthPsychology & healthPsychol HealthPsychology & healthPsychol HealthPsychology & health1113-27269*BehaviorCharacter*Decision MakingEmotionsHabitsHumansIndividuality*IntentionInternal-External Control*MotivationProblem Solving*Psychological TheorySocial Conformity2011Sep1476-8321 (Electronic)
0887-0446 (Linking)21929476http://www.ncbi.nlm.nih.gov/pubmed/2192947610.1080/08870446.2011.613995( HYPERLINK \l "_ENREF_3" \o "Ajzen, 2011 #994" Ajzen, 2011). Normative beliefs were often measured with very brief scales, so it may be that some null findings were due to measurement error. The positive relationship between perceived accessibility and knowledge of testing sites, particularly in resource limited contexts, are consistent with global efforts to increase HIV testing availability ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_31" \o "Iwuji, 2016 #1165" Iwuji et al., 2016).
There was minimal evidence of a relationship between fear of testing and testing intention, in contrast to findings between greater fear and reduced testing ADDIN EN.CITE Evangeli2016998(Evangeli et al., 2016)99899817Evangeli, M.Pady, K.Wroe, A. L.Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK. michael.evangeli@rhul.ac.uk.
Department of Psychology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK.Which Psychological Factors are Related to HIV Testing? A Quantitative Systematic Review of Global StudiesAIDS BehavAIDS and behaviorAIDS BehavAIDS and behaviorAIDS BehavAIDS and behavior880-9182042016Apr1573-3254 (Electronic)
1090-7165 (Linking)26566783http://www.ncbi.nlm.nih.gov/pubmed/26566783479926710.1007/s10461-015-1246-0( HYPERLINK \l "_ENREF_18" \o "Evangeli, 2016 #998" Evangeli et al., 2016). There may have been measurement error associated with the few items used to measure testing fear. It may also be that testing fear does not affect testing intention but that enacting this motivation (i.e., actually testing) is difficult in the presence of high levels of fear. Perceiving oneself to be at risk of HIV was consistently associated with testing intention across a range of contexts. Intending to test may be a functional way of coping with perceived HIV risk, consistent with models of stress and coping ADDIN EN.CITE Folkman19861181(Folkman, Lazarus, Gruen, & DeLongis, 1986)1181118117Folkman, S.Lazarus, R. S.Gruen, R. J.DeLongis, A.Appraisal, coping, health status, and psychological symptomsJ Pers Soc PsycholJournal of personality and social psychologyJ Pers Soc PsycholJournal of personality and social psychologyJ Pers Soc PsycholJournal of personality and social psychology571-9503*Adaptation, PsychologicalAdjustment Disorders/*psychologyAdultFemaleHumans*Life Change EventsMaleMiddle AgedPersonality TestsPsychometricsSomatoform Disorders/*psychology1986Mar0022-3514 (Print)
0022-3514 (Linking)3701593http://www.ncbi.nlm.nih.gov/pubmed/3701593( HYPERLINK \l "_ENREF_22" \o "Folkman, 1986 #1181" Folkman, Lazarus, Gruen, & DeLongis, 1986).
Limitations
Given the small number of studies, we relied on crude number counts to make inferences about the overall strength of associations. Future reviews should be able to conduct meta-analyses, along with assessing moderators of relationships (e.g., sex, study location, previous testing and testing context). There was an absence of prospective cohort designs. These would have allowed more confident causal inferences about the relationship between potential psychological determinants and testing intention. Experimental designs, potentially in the context of intervention studies, were absent. HIV testing intervention studies measuring effects on psychological determinants of testing and testing itself would be consistent with guidelines stressing the need to assess how complex interventions work ADDIN EN.CITE Craig20081189(Craig et al., 2008)1189118917Craig, P.Dieppe, P.Macintyre, S.Michie, S.Nazareth, I.Petticrew, M.Medical Research Council, GuidanceMRC Population Health Sciences Research Network, Glasgow G12 8RZ. peter@sphsu.mrc.ac.ukDeveloping and evaluating complex interventions: the new Medical Research Council guidanceBMJBmjBMJBmjBMJBmja1655337*Diffusion of InnovationEvaluation Studies as TopicEvidence-Based Medicine*Practice Guidelines as TopicTherapeutics20081756-1833 (Electronic)
0959-535X (Linking)18824488http://www.ncbi.nlm.nih.gov/pubmed/18824488276903210.1136/bmj.a1655( HYPERLINK \l "_ENREF_12" \o "Craig, 2008 #1189" Craig et al., 2008).
There were some potentially important psychological variables rarely assessed. Cues for testing and support for testing were only included in predictors once, both in the same study ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_59" \o "Westmaas, 2012 #1069" Westmaas et al., 2012). Studies varied on measures used and their reliability and validity. In particular, HIV testing intention was assessed inconsistently, frequently with single items with a yes/no response option. It is unlikely that this represents the complexity of the construct. We defined HIV intention broadly, including studies measuring willingness to test. Although these outcomes are related, it may be that there are conceptual differences that explain variations in findings. Finally, only five of the twenty studies met our external validity criteria and only five met our internal validity criteria, despite the latter, in particular, being generous.
Theoretical, research and practice implications
When models were tested in their entirety, they only explained a relatively small proportion of the variance in testing intention ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_1" \o "Abamecha, 2013 #1000" Abamecha, Godesso, & Girma, 2013). In addition, although some variables frequently cited as intention determinants were consistently associated with testing intention (e.g., pro-testing attitudes, perceived risk), others were not (e.g., perceived HIV severity, social norms, testing barriers). Furthermore, some variables that do not appear explicitly in theoretical models applied to this context did appear to show consistent relationships with testing intention (e.g., perceived testing availability). It seems clear that existing models do not adequately explain HIV testing intention determinants.
There is a need to clearly differentiate current and future perceived risk as they showed different relationships with testing intention. This goal would be facilitated by developing reliable and valid measures of perceived HIV risk. More generally, using or developing robust measures of both potential psychological determinants and HIV testing intention itself, is a priority. In addition, it would be useful to assess some rarely measured variables, for example, well-being, and emotions such as shame and guilt relating to testing. In general, emotions were rarely included in studies. The differing findings for positive and negative attitudes suggests that it is important to examine these separately in future studies. It will also be worthwhile examining the relationship between testing barriers and testing intention in Africa as there were no studies where this had been done.
We suggest practice implications for enhancing HIV testing intention (and potentially HIV testing itself) cautiously given the nature of the evidence from this review. Nevertheless, findings suggest that it may be fruitful to focus on strategies to emphasise the benefits of testing, and to both provide information on testing sites/formats and on increasing accessibility to testing. In addition, efforts to increase the personal relevance of HIV risk may contribute to a stronger intention to test. Such interventions can be delivered at a number of levels: individual ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_4" \o "Alemagno, 2009 #1191" Alemagno, Stephens, Stephens, Shaffer-King, & White, 2009), social/relational ADDIN EN.CITE Van Rompay20081192(Van Rompay et al., 2008)1192119217Van Rompay, K. K.Madhivanan, P.Rafiq, M.Krupp, K.Chakrapani, V.Selvam, D.Sahaya International Inc., Davis, USA. kkvanrompay@ucdavis.eduEmpowering the people: development of an HIV peer education model for low literacy rural communities in IndiaHum Resour HealthHuman resources for healthHum Resour HealthHuman resources for healthHum Resour HealthHuman resources for health6620081478-4491 (Electronic)
1478-4491 (Linking)18423006http://www.ncbi.nlm.nih.gov/pubmed/18423006237724910.1186/1478-4491-6-6( HYPERLINK \l "_ENREF_56" \o "Van Rompay, 2008 #1192" Van Rompay et al., 2008), population ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_37" \o "Kaufman, 2014 #1196" Kaufman et al., 2014), and structural ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_31" \o "Iwuji, 2016 #1165" Iwuji et al., 2016).
Figure 1: Study search process
External validityRepresentativeness of the sample for the target populationWas the sample a consecutive or random sample, or were all of the population eligible? The percentage of selected individuals who agreed to participate and were included in the analysis
Were at least 80% of those eligible to participate included in the final analysis?Internal validityPerformance bias 1. Measurement of testing intentionWas testing intention measured with more than one item or with a likert scale?Detection bias 2. Measurement of psychological factorsWas there some evidence that psychological variables were measured reliably and validly? Selection bias/control of confounding variable (systematic bias associated with comparison group)Extent to which possible confounding variables were measured and analysed appropriately
Were possible confounding variables (a) measured (b) considered in the analysis?
ReferenceLocation, inclusion/
exclusion criteria and context
Design, sampling and sample
Measurement of psychological factors
(using or adapting existing scales if stated)
Measurement of HIV testing intention/
readiness/
willingnessAssociations between psychological factors and HIV testing intention
Abamecha et al ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_1" \o "Abamecha, 2013 #1000" Abamecha et al., 2013)Ethiopia.
Health Professionals working in public sector, with at least diploma level education.
.
Cross-sectional. Random sampling.
n= 336 (121 female). Response rate 91%. Mean age 27.5 years.
Attitude towards testing Direct: 4 items, bipolar differential scale; Indirect: 6 items, likert scale
Subjective Norms Direct: 4 item bipolar differential scale; Indirect: 6 item Likert Scale.
Perceived Behavioral Control (PBC) Direct: 4 item bipolar differential scale; Indirect: 4 item Likert scale.
Perceived Risk (current) 5 item, Likert scale; 1 dichotomous item.
Perceived Severity - 5 item, Likert scale; 1 dichotomous item.
Self-report questionnaire. Scales adapted and modified from two previous studies ADDIN EN.CITE ADDIN EN.CITE.DATA ( HYPERLINK \l "_ENREF_21" \o "Fishbein, 2005 #962" Fishbein & Ajzen, 2005; HYPERLINK \l "_ENREF_48" \o "Pikard, 2009 #1074" Pikard, 2009)Intention to test
4 item, Likert scale.Attitude towards testing more positive Direct Attitude related to greater intention ( = 0 . 1 8 , p <