Unlocking Digital Health: Inequalities in the adoption of a Patient Portal

ObjectiveDigital health apps and patient portals are proposed as part of the drive from analogue to digital care for the NHS 10 Year Plan. Without mitigation strategies, digital inequalities could arise as a result and more evidence is needed to understand how to mitigate this. MethodsAs part of an equalities impact assessment, a retrospective cross-sectional analysis was conducted examining patient portal activation among patients invited to outpatient appointments at two large south-east London Hospital Trusts between May 1st and November 1st, 2024. Results503,688 patients invited to attend outpatients during the study period, 52.7% of patients invited to attend outpatients had activated the patient portal. Availability of email contact details were strongest determinant for likelihood of onboarding (OR 10.86; CI 10.60-11.12). Multivariate logistic regression models showed the following groups were less likely to activate the patient portal. Men (odds ratio 0.84 (CI:0.83-0.85), extremes of age (71-80 years or 11-20 years), those mixed or undefined ethnicity OR 0.58 (CI 0.57- 0.59), black ethnicities OR 0.62 (CI 0.61-0.64) or un-recorded ethnicities OR 0.72 (CI 0.7- 0.74) and those with highest degree of socio-economically deprivation (IMD group 1) OR 0.68 (CI 0.65-0.72). ConclusionThis large scale roll-out of a digital health portal provide empirical evidence of factors which drive digital inequalities for patients of two major London NHS Trusts. The observed disparities across demographic and socioeconomic dimensions and simple reliable digital contact mechanisms highlight the risk that digital healthcare initiatives may inadvertently produce new types of inequalities. What is the paper about?O_LIWere there inequalities in activation of the patient portal, MyChart, in the Apollo programme, by demographic characteristics of the patients? Yes. Table 1 C_LIO_LIWere the observed inequalities attributable to confounding? No. In a multivariate logistic regression, the inequalities persisted across all variables. Figure 1 C_LIO_LICould the inequalities be explained by differential access to email and mobile phones across the groups? For several variables, adjusting for the presence of email address and mobile phone number attenuated the strength of the relationship with Patient Portal activation. It did not remove the effect for any variable and the relationship with ethnicity was barely affected. Figure 2 C_LI O_TBL View this table: org.highwire.dtl.DTLVardef@62ce0aorg.highwire.dtl.DTLVardef@c4f9e6org.highwire.dtl.DTLVardef@1f2a8fborg.highwire.dtl.DTLVardef@341db1org.highwire.dtl.DTLVardef@1896415_HPS_FORMAT_FIGEXP M_TBL O_FLOATNOTable 1:C_FLOATNO O_TABLECAPTIONPatient portal activation by Sex, Age group, Ethnic Group, Index of multiple deprivation, availability of email address and mobile phone number and first language among 499,098 patients invited to outpatient clinics across two London teaching hospital systems in south east London, UK, in 2024. C_TABLECAPTION C_TBL O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=145 SRC="FIGDIR/small/25330473v2_fig1.gif" ALT="Figure 1"> View larger version (23K): org.highwire.dtl.DTLVardef@109dafcorg.highwire.dtl.DTLVardef@7de1aorg.highwire.dtl.DTLVardef@86a602org.highwire.dtl.DTLVardef@1fce83e_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOFigure 1:C_FLOATNO Odds Ratios for adoption of Patient Portal by Sex, Age Group, Ethnic Group and Index of Multiple Deprivation among 499,098 patients invited to outpatients across two London teaching hospital systems between 26 May 2024 and 26 November 2024. Multivariate Logistic Regression. C_FIG O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=145 SRC="FIGDIR/small/25330473v2_fig2.gif" ALT="Figure 2"> View larger version (24K): org.highwire.dtl.DTLVardef@f379eforg.highwire.dtl.DTLVardef@ca6ba3org.highwire.dtl.DTLVardef@160a2f1org.highwire.dtl.DTLVardef@169aa0e_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOFigure 2:C_FLOATNO Odds Ratios for adoption of a patient portal, by Sex, Age Group, Ethnic Group and Index of Multiple Deprivation, before and after adjusting for the presence of an email address and mobile phone number C_FIG