The association between stress and episodes of symptom manifestation of genital herpes: a systematic review of the literature Eamann Breen The Mortimer Market Centre "No article that succeeds in being published is accompanied by a guarantee of excellence" (Meltzoff). “It is more important to know what kind of person has the disease than what kind of disease the person has”. (Hippocrates) Background Objectives Search strategy Selection criteria Data collection and analysis Main results Limitations of review Conclusions Genital Herpes Pathology Burden of Disease What is stress? Definitions Measurements of stress Causes and triggers Stress and the immune system The effect of stress on the PNI system Cause Immune Effect Increase in system/endocrine Virus reactivation stress system Psychological well being Aims and Objectives The aim of this study is to investigate systematically the hypothesis that there is a relationship between an increase in stress and recurrent genital herpes, specifically where stress acts as a predictor for repeat episodes of the virus. Whether there is an association between patient psychological stress and recurrent episodes of genital herpes? How reliable are the existing studies? Where is the potential for further research? Definition A Systematic review “is a review of a clearly formulated question that uses systematic and explicit methods to identify, select and critically appraise relevant research, and to collect and analyse data from studies that are included in the review." Why undertake a systematic review? Research into what is perceived by patients as stressful and what leads to recurrence has already been undertaken. Additional primary research was rejected with a view to review the work already available. The purpose of this research is to identify information that may not be apparent from individual studies where the effects under investigation may be small, or where there are conflicting findings from different studies. Sources of published and ongoing systematic reviews Databases The Cochrane Database of Systematic Reviews (CDSR) Database of Abstracts of Reviews of Effectiveness (DARE) Health Technology Assessment (HTA) Database NHS Economic Evaluation Database NHSEED Internet Sites TRIP - http://www.tripdatabase.com Health services/technology assessment text (HSTAT) - http://text.nim.nih.gov/ National Coordinating Centre for Health Technology Assessment - http://www.hta.nhsweb.nhs.uk/ NICE appraisals - http://nice.org.uk/nice- web/cat.asp?c=153 Research in progress National Research Register (NRR) - http://www.update- software.com/national/nrr-frame.html Adapted from (Glanville and Sowden 2001) Guidelines for establishing causation Temporal Relation Does cause precede the effect Plausibility Is the association consistent with other knowledge Consistency Have similar results been shown in other studies Strength What is the strength of the association Dose-response Is variance associated with cause and effect relationship Reversibility Does removal of possible cause lead to reduction of disease risk Study Design Is the evidence based on strong study design Judging the evidence How many lines of evidence lead to that conclusion Adapted from (Beaglehole, Bonita et al. 1993) Study Selection adapted from (Khan and Kleijnen 2001) Review Question Is there a relationship between stress and recurrent genital herpes? Selection Criteria Inclusion Criteria Exclusion Criteria Population Patients diagnosed with No proven diagnosis recurrent herpes Other conditions Measures Specific measures of stress Other psychological conditions Study Design Cohort studies Case/control studies Descriptive case studies RCT's Reviews Publication PubMed Not published in English Embase 1980 to present prior to 1980 Grading of Review Studies Grade Level of Effectiveness Test Accuracy Efficiency Evidence A 1 High quality High quality studies with a High quality evaluations of important experimental blind comparison of test to alternative interventions comparing all studies without reference standard in an relevant outcomes against appropriate heterogeneity appropriate population cost measurement, including a sensitivity and with precise spectrum analysis results B 2/3 Low quality Any one or two of the Evaluations without relevant outcomes experimental following: or without appropriate cost measurement studies, high Narrow population quality spectrum controlled Differential use of observational reference standard studies Reference standard not blind Case control study C 4 Low quality Any three or more of the Evaluations without sensible sensitivity controlled above analysis observational studies, case series D 5 Expert opinion Expert opinion Expert opinion Adapted from (Sackett, Straus et al. 2000); (Khan and Kleijnen 2001) Study Search Potentially relevant citations identified after liberal screening of the electronic search (n=64) Citations excluded with reasons (n= 32) after evaluation Studies retrieved for more detailed evaluation (n=32) Studies excluded (after evaluation of full text) from systematic review with reasons (n=16) Relevant studies included within systematic review (n=16) Adapted from (Khan and Kleijnen 2001) Results Included studies Excluded studies Prospective and retrospective studies Study/Design Subjects Timescale Findings Grade Level of evidence (Cassidy, Meadows et al. 1997) N = 116, m= 48, f =68 n/a No relationship B 3 UK, Retrospective (Swanson, Dibble et al. 1995) N = 70, m = 41%, F = 59% n/a supports relationship B 3 US, Retrospective (Swanson and Dibble 1993) N = 252, m = 59, f = 193 n/a supports relationship C 4 US, Retrospective (Brooks, Haywood et al. 1993) N = 90, 40 =m, 50 = f, n/a supports relationship B 3 UK, Retrospective median age = 34, range 19 - 60 (Keller, Jadack et al. 1991) N = 60, m =26, f =34, n/a No relationship C 4 US, Retrospective mean age 31.7 (S.D =/- 8.1) (Longo and Clum 1989) N = 46, m = 18, f = 28 n/a supports relationship B 3 US, Retrospective (VanderPlate and Kerrick 1985) N = 59, m =16, w = 43. n/a Contradictory relationship B 3 US, Retrospective (Silver, Auerbach et al. 1986) N = 67, m =32, f = 35, n/a supports relationship B 3 US, Retrospective mean age = 33 range 20 - 65 (Bierman 1983) N = 375, m = 202, f = 173 n/a supports relationship C 4 US, Retrospective (Cohen, Kemeny et al. 1999) N = 58, all f, 6 months supports relationship B 3 US, Prospective mean age = 31.8 (S.D +/- 5.8) (Kemeny, Zegans et al. 1987) N = 36 f = 30, m = 6, 6 months No relationship B 3 US, Prospective mean age 33.4, (SD 9.5, range 18 -69) (Rand, Hoon et al. 1990) N = 64, m = 13, f = 5, 6 months No relationship B 3 US, Prospective mean age =27.9 (S.D +/- 5.7) Intervention Studies Study Subjects Intervention Measures (VanderPlate and N = 4, all f Biofeedback, and 2 subjects given Electromyographic (EMG) biofeedback, Kerrick 1985) US Muscle relaxation treatment 2 subjects given progressive muscle relaxation treatment (PT) Timescale 6 months Results Biofeedback resulted in a 72% and 7% decrease in recurrences, PT resulted in 66% and 100% reduction in recurrences Grade B Level of Evidence 3 Longo, Clum et al. 1998) N = 31 not specified Psychosocial intervention, Profile of Mood States US social support, and Zung Depression Scale waiting list control UCLA Loneliness Scale MHLC scales Hassles Scale and Contact Questionnaires. Timescale 6 months Results Perceived stress did not differ among treatment conditions Psychosocial interventions reduced the severity of recurrences Grade B Level of Evidence 3 Burnette, Koehn N = 8, all f Progressive muscle relaxation treatment Questionnaires et al. 1991) US Timescale 6 months Results 5/8 participants reported a significant reduction in recurrence rate Grade B Level of Evidence 3 (Koehn, Burnette N = 4, all f Applied relaxation skills Daily diary et al. 1993) US Pre and post treatment relaxation evaluations Pre and post treatment relaxation application evaluations Treatment satisfaction EMG measures Timescale 3 months Results All subjects reported post treatment reductions in herpes frequency. Statistically significant for 3/4 participants Grade B Level of Evidence 3 Excluded Studies Sixteen studies were excluded from the review. The excluded studies fit into four categories no specific measures of stress (Pederson and Stavraky 1987); (Hillard, Hillard et al. 1989); (Manne and Sandler 1984); (Luby and Klinge 1985); (Schofield, Minichiello et al. 2000); (Jadack, Keller et al. 1991); (Stout and Bloom 1986) case studies (Brown and Callen 1999); (Gould and Tissler 1984); (Lovejoy 1987) secondary analysis (Dibble and Swanson 2000); (Kemeny, Cohen et al. 1989) original scores not included (Derman 1986); (Guinan, MacCalman et al. 1981); (Hoon, Hoon et al. 1991) Gender The majority of participants for all the studies were women with some studies made up entirely of women. No study indicated the sexual orientation of the participants. Sample size Sample size across all designs varied greatly, from N = 4 (VanderPlate and Kerrick 1985) to 375 (Bierman 1983). The median range for the prospective and retrospective studies was between 60 - 70 participants. Measures of stress Measures of stress differed from study to study with some studies using a variety of different and accepted methods. Some used subjective methods. Others used a combination of questionnaires and medical tests. One study used CD4 and CD8 counts and one used lesion culture. Health Behaviours No record of health enhancing and impairing behaviours, such as diet, exercise, alcohol and tobacco use, which can compromise or benefit health were indicated as being measured in any of the studies. Recruitment of participants Patients were recruited from different sources ranging from GU clinic to newspaper advertising or both. Age Age range where given also varied from 19 - 69 years. Mean age varied where indicated but tended to be in the early 30‟s All of this indicates that in general the samples were not representative, but with the lack of further information, we could assume that this was representative of the local population diagnosed with genital herpes. Principle Findings of the Review The evidence suggests that there is a weak relationship between stress and recurrent genital herpes. Participants in the retrospective studies report that they experience increased stress prior to latent virus reactivation. However the results of the prospective studies do not support this conclusion, but one concludes that persistent stress can lead to reactivation. The four intervention studies demonstrate that stress management training lead to decreased stress and recurrences and these studies represent the strongest evidence that links stress and recurrent herpes. A comparison of the studies was complicated by the differences in design and studies that included interventions and by the paucity of research in this field. Limitations of the Review This systematic review is based on work designed and completed by other researchers and the strengths and weaknesses of these studies all have an impact on the findings of the review. It is impossible to be certain that all available research has been discovered but in attempting to appraise and offer a critical analysis of what has been available it could be said that this review has highlighted issues mentioned by previous research in this field. Inclusion of studies from the last twenty years only was intended to exclude the majority of commentary/expert opinion type papers that existed up to this time. The use of papers published in English only may again be another form of bias. Most studies were undertaken in the US a resource rich country with ample resources for research and where HSV is considered to be a major public health concern. The following factors may have an impact on the results and the internal validity of the primary studies; History History X Assignment Maturation Testing can have effects of its own. Instrumentation/Raters. Causation Table Stress and recurrent Genital Herpes Temporal Relation The cause does precede effect, but not always Plausibility The association is consistent with other knowledge Consistency Some consistency Strength Medium Dose-response relationship Relationship does exist Reversibility Difficult to be clear about reversibility Study Design Weak to medium study design Judging the evidence Limited evidence currently exists Adapted from (Beaglehole, Bonita et al. 1993) Model for further research Information Required Information at at Baseline 0,1,2,3,4,5,6 months Sex CD4 counts High frequency recurrers Age CD8 counts Social Class NK counts Low frequency recurrers Months since diagnosis Frequency of recurrence Treatment used Stress measures/ stress measures questionnaires Implications of research for practice What advice are we providing? What is the purpose of this advice? Do we have to give „positive‟advice? Conclusions The relationship between stress and recurrent genital herpes is complicated. The results of this systematic review suggest that there is a relationship between stress and recurrent herpes however the precise nature of this relationship needs to explored further. The relationship has not been established as a causal one.