Psychological interventions for people with hemophilia

Managing hemophilia is challenging both in terms of medical treatment and its broad impact on many aspects of the individual's life, including self-perception. Several psychosocial issues are potentially relevant in the clinical management of hemophilia, including it being a chronic and incurable condition; e.g. people with hemophilia must adapt to optimally interact with peers and to practice sports - even choosing a sport represents an issue for perceived limitations, expectations and cultural influences on the individual and their family. People with hemophilia can react by denying their condition and its manifestations and not adhering to treatment. Due to the complexity of relationships surrounding genetic diseases, parents and relatives may have their own issues that contribute to making life easier or more difficult for the person with hemophilia. Anxiety, sadness and depression resulting in mental health disorders are reported in this population and may influence quality of life (QoL) depending on cultural background, religious beliefs, family support and other variables.

Primarily to assess the effectiveness of psychological therapies for improving the ability of people with hemophilia to cope with their chronic condition.

We aimed to identify trials from the Cochrane Cystic Fibrosis and Genetic Disorders Group's Coagulopathies Trials Register, Embase and PsycINFO, CINAHL, MEDLINE and trial registries. We searched reference lists of included publications. Most recent search of the Group's register: 13 June 2019.

Randomized controlled trials (RCTs) and quasi-RCTs in people with hemophilia of any age or gender, type A or B, any severity, with or without inhibitors, with or without HIV or hepatitis C virus. All psychological interventions for promoting emotional, intellectual and spiritual wellness. Individual, group or family group therapy interventions were eligible.

We independently assessed trials, extracted data and assessed the risk of bias and assessed the quality of the evidence using GRADE.

Seven trials were included (362 participants randomized, data from 264 participants available for analysis); six of parallel design and one a partial cross-over design. One multicenter trial was conducted in Canada; the remaining six were single centre undertaken in the UK, USA, Iran and in the Netherlands. All trials had a high risk of bias for participant blinding and use of patient-reported outcomes. Evidence was retrieved on four interventions: psycho-education (DVD plus information booklet versus information booklet alone; computerised learning versus no intervention); cognitive therapy (auto-hypnosis (self-hypnosis) versus control); and behavioural therapy (relaxation (progressive or self control) versus no treatment). We also aimed to assess psychodynamic therapy and systemic therapy, but no trials were identified. Heterogeneity of the outcome measures and measurements precluded meta-analyses. No trial reported the cost of the psychological intervention and family adjustment. DVD plus information booklet compared to information booklet alone One trial (108 participants) showed coping strategies may lower pre-contemplation scores and negative thoughts, mean difference (MD) -0.24 (95%CI -0.48 - 0.00, low-certainty evidence), however, other measures of coping strategies in the same trial suggest little or no difference between groups, e.g. contemplation, MD (-0.09, 95%CI -0.32 - 0.14, low-certainty evidence). The same trial measured QoL and showed little or no difference between treatment groups for the physical domain, MD 0.59 (95% CI -3.66 to 4.84, low-certainty evidence), but may improve scores in the mental health domain for those receiving the booklet plus DVD compared to booklet alone, MD (4.70, 95% CI 0.33 to 9.07, low-certainty evidence). Mood or personal well-being were not reported. Computerised learning compared to no intervention Two trials (57 participants) reported on interventions aimed at children and adolescents and their impact on promoting a sense of self-efficacy (primary outcome 'Mood and personal well-being'), but only one showed an increase, MD 7.46 (95%CI 3.21 to 11.71, 17 participants, very low-certainty evidence); the second did not report control group data. One trial (30 participants) showed the intervention did not improve self-efficacy in adults, but appropriate data could not be extracted. Two trials (47 participants) reported coping strategies; one only reported within-group differences from baseline, the second showed an increase from baseline in coping strategies in the Internet program group compared to the no intervention group (disease-specific knowledge, MD 2.45 (95% CI 0.89 to 4.01); self-management ability and transition readiness, MD 19.90 (95% CI 3.61 to 36.19; low-certainty evidence). One trial reported QoL but with insufficient information to calculate changes from baseline; no difference in post-treatment scores was seen between groups, MD -8.65, 95% CI -18.30 to 1.00, very low-certainty evidence). Auto-hypnosis (self-hypnosis) compared to control There were two older trials that reported on this intervention (50 participants) focusing mainly on the secondary outcome 'physical health'; only one trial reported the primary outcome 'mood and personal well-being' (only within-group differences in the treatment group). Coping strategies and QoL were not assessed in the trials. Relaxation (progressive or self control) compared to no treatment Only one trial (seven participants) from 1985, was included which focused on 'physical health' and did not report on any of our primary outcomes.

Not all of the seven included trials analysed the effects of the interventions on our primary outcomes (mood and personal well-being, coping strategies and QoL). Three trials were conducted in the 1970s and 1980s using techniques of auto-hypnosis or relaxation and, in accordance with the needs and therapeutic possibilities of the time, they focused on secondary outcomes, e.g. frequency of bleeding (physical health) and adherence to the intervention. The four newer trials assessed psycho-educational interventions all mediated by the use of technologies (DVD or computer) and often created according to age needs of the target group. In these cases, attention was shifted to our pre-defined primary outcomes. This review has identified low- and very low-certainty evidence, prompting caution in its interpretation. The major problem we encountered was the heterogeneity of trial designs, of interventions and of outcome measures used across the trials. We strongly suggest that researchers consider developing a core outcome set to streamline future research; randomization was proven to be safe and acceptable, and blinding should be considered for those assessing patient-reported outcomes.