Health Equity in the Effectiveness of Web-Based Health Interventions for the Self-Care of People With Chronic Health Conditions: Systematic Review

Abstract

Background: Web-based self-care interventions have the potential to reduce health inequalities by removing barriers to access to health care. However, there is a lack of evidence about the equalizing effects of these interventions on chronic conditions.

Objective: This study investigated the differences in the effectiveness of web-based behavioral change interventions for the self-care of high burden chronic health conditions (eg, asthma, chronic obstructive pulmonary disease [COPD], diabetes, and osteoarthritis) across socioeconomic and cultural groups.

Methods: A systematic review was conducted, following Cochrane review guidelines. We conducted searches in Ovid Medical Literature Analysis and Retrieval System Online and Cumulative Index to Nursing and Allied Health Literature databases. Studies with any quantitative design were included (published between January 1, 2006, and February 20, 2019) if they investigated web-based self-care interventions targeting asthma, COPD, diabetes, and osteoarthritis; were conducted in any high-income country; and reported variations in health, behavior, or psychosocial outcomes across social groups. Study outcomes were investigated for heterogeneity, and the possibility of a meta-analysis was explored. A narrative synthesis was provided together with a novel figure that was developed for this review, displaying heterogeneous outcomes.

Results: Overall, 7346 records were screened and 18 studies were included, most of which had a high or critical risk of bias. Important study features and essential data were often not reported. The meta-analysis was not possible due to the heterogeneity of outcomes. There was evidence that intervention effectiveness was modified by participants' social characteristics. Minority ethnic groups were found to benefit more from interventions than majority ethnic groups. Single studies with variable quality showed that those with higher education, who were employed, and adolescents with divorced parents benefited more from interventions. The evidence for differences by age, gender, and health literacy was conflicting (eg, in some instances, older people benefited more, and in others, younger people benefited more). There was no evidence of differences in income, numeracy, or household size.

Conclusions: There was evidence that web-based self-care interventions for chronic conditions can be advantageous for some social groups (ie, minority ethnic groups, adolescents with divorced parents) and disadvantageous for other (ie, low education, unemployed) social groups who have historically experienced health inequity. However, these findings should be treated with caution as most of the evidence came from a small number of low-quality studies. The findings for gender and health literacy were mixed across studies on diabetes, and the findings for age were mixed across studies on asthma, COPD, and diabetes. There was no evidence that income, numeracy, or the number of people living in the household modified intervention effectiveness. We conclude that there appear to be interaction effects, which warrant exploration in future research, and recommend a priori consideration of the predicted interaction effects.

Trial registration: PROSPERO CRD42017056163; https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=56163.

Keywords: asthma; diabetes; eHealth; health equity; intervention; osteoarthritis; pulmonary disease, chronic obstructive; self-care.

Figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analysis flow chart.

Figure 2

Risk of bias table for randomized controlled trials.

Figure 3

Risk of bias assessment for the nonrandomized controlled trial (RCT) studies. ^aOverall risk of bias: equal to the most severe level of bias found in any domain.

Figure 4

Adapted Harvest plot with evidence for the modification of intervention effect by increasing age for chronic obstructive pulmonary disorder studies. RCT: randomized controlled trial.

Figure 5

Adapted Harvest plot with evidence for increasing age-modifying intervention effects across outcomes in diabetes studies. RCT: randomized controlled trial.

Figure 6

Adapted Harvest plot for gender-modifying intervention effect across outcomes in diabetes studies. RCT: randomized controlled trial.

Figure 7

Adapted Harvest plot for higher education modifying intervention effects across outcomes in diabetes studies. RCT: randomized controlled trial.

Figure 8

Adapted Harvest plot for minority ethnic group modifying intervention effects across outcomes in diabetes studies. RCT: randomized controlled trial.

Figure 9

Adapted Harvest plot for higher health literacy modifying intervention effects across outcomes in diabetes studies. RCT: randomized controlled trial.