Disparities in healthcare access and utilization among people living with HIV in China: a scoping review and meta-analysis

Background Healthcare disparities hinder the goal of ending the HIV pandemic by 2030. This review aimed to understand the status of healthcare disparities among people living with HIV (PLWH) in China and summarize driving factors. Methods We searched six databases: PubMed, Web of Science, Cochrane Library, Scopus, China National Knowledge Infrastructure (CNKI), and China Wanfang. English or Chinese articles published between January 2000 and July 2022 were included if they focused on any disparities in access to and utilization of healthcare among PLWH in China. Grey literature, reviews, conferences, and commentaries were excluded. A random effects model was used to calculate the pooled estimates of data on healthcare access/utilization and identified the driving factors of healthcare disparities based on a socio-ecological framework. Results A total of 8728 articles were identified in the initial search. Fifty-one articles met the inclusion criteria. Of these studies, 37 studies reported HIV-focused care, and 14 focused on non-HIV-focused care. PLWH aged ≥ 45 years, female, ethnic minority, and infected with HIV through sexual transmission had a higher rate of receiving antiretroviral therapy (ART). Females living with HIV have higher adherence to ART than males. Notably, 20% (95% CI, 9–43%, I2 = 96%) of PLWH with illness in two weeks did not seek treatment, and 30% (95% CI, 12–74%, I2 = 90%) refused hospitalization when needed. Barriers to HIV-focused care included the lack of knowledge of HIV/ART and treatment side effects at the individual level, and social discrimination and physician-patient relationships at the community/social level. Structural barriers included out-of-pocket medical costs, and distance and transportation issues. The most frequently reported barriers to non-HIV-focused care were financial constraints and the perceived need for medical services at individual-level factors; and discrimination from healthcare providers, distrust of healthcare services at the community/social level. Conclusion This review suggests disparities in ART access, adherence, and utilization of non-HIV-focused care among PLWH. Financial issues and social discrimination were prominent reasons for healthcare disparities in PLWH care. Creating a supportive social environment and expanding insurance policies, like covering more medical services and increasing reimbursement rates could be considered to promote healthcare equity.

for resolving healthcare disparities in China.
The aims of this study are to 1) understand the status of healthcare disparities among PLWH in China in both HIV-focused and non-HIV-focused care, and 2) summarize the driving forces of healthcare disparities among PLWH in China.

Study design
A scoping review approach was adopted as the research method because it allows researchers to rapidly map out key concepts in a complex research area well based on the main sources of evidence available [17]. The July 2022 and published in English or Chinese were eligible for inclusion. All identi ed articles from the searches were transferred to the EndnoteX9 software tool for managing bibliographies, citations, references, and sharing with other reviewers. Duplicate articles were excluded.

Eligibility Criteria
All included studies met the following criteria: 1) the study had to be conducted among people living with HIV in China, 2) focused on disparities in health care, including access and utilization of medical services, and 3) used quantitative, qualitative, or mixed method study designs. To keep the inclusion criteria as broad as possible, dissertations were eligible. However, grey literature, reviews, conference articles, and commentaries were excluded.

Data Screening and Extraction
Two reviewers (WA and CF) rst independently assessed the titles and abstracts to identify relevant records for inclusion following the eligibility criteria. When two reviewers disagreed on article selection, a consensus was sought from the third researcher (WT). Full texts of the eligible studies were retrieved and assessed for inclusion following the same screening method. The following data were recorded from the nal included studies independently by the same reviewers using Excel data extraction and synthesis sheets. As data were extracted, a data charting form was collectively developed to map studies based on forms of disparity, and the extracted data included, author name (s), year, region, study population, sample sizes, study design, factors associated with healthcare access and utilization (like types of healthcare, variables, and data), and facilitators and barriers of healthcare disparities.

Data Analysis
Quantitative data on health care access/utilization among PLWH were pooled using a random-effects model in a meta-analysis utilizing R-Language Version 4.1.1. The rate of receiving ART was de ned as the number of people receiving ART in his/her survey period divided by the total number of people surveyed. Full ART adherence was de ned as participants not missing a dose of antiretroviral drugs in his/her survey period. The 95% and above ART adherence was de ned as participants taking ≥ 95% of the dose of antiretroviral drugs in his/her survey period. Higher than 90% ART adherence was de ned as participants taking > 90% of the dose of antiretroviral drugs in his/her survey period. The two-week morbidity rate was de ned as the number of persons who had a disease in the past two weeks before his/her survey divided by the total number of people surveyed [19]. The two-week visit rate was de ned as the number of people who utilized outpatient services in the past two weeks before his/her survey divided by the total number of people surveyed [19]. The non-visit rate of people with illnesses in two weeks was de ned as the number of people with illnesses but without medical consultation in the past two weeks before his/her survey divided by the number of people with illnesses in the past two weeks before his/her survey [19]. The annual inpatient rate was de ned as the number of people hospitalized in the past year before his/her survey divided by the total number of people surveyed [19]. The rate of non-hospitalization among those who required hospitalization was de ned as the number of people who required hospitalization but were not hospitalized divided by the number who required hospitalization [19]. In addition, the driving factors of healthcare disparities were classi ed into three categories based on a socio-ecological framework: individual level, community and interpersonal level, and structural level [20]. Individual-level factors refer to biological and other personal characteristics, such as knowledge, behavior, income, health history, etc. Community and social level factors are characterized as related to the individual's social networks and life environment. Structural factors include cultural and social norms and health, economic and social policies.

Results
Overall characteristics of the included studies Figure 1 is the PRISMA ow diagram. The initial search generated a total of 8728 articles. After excluding duplicate references, 7291 articles remained. A review of titles and abstracts revealed that 6669 articles were irrelevant or not conducted in China, leaving 622 articles for full-text screening. Of these, 571 articles did not meet the inclusion criteria because they were review/ comments (n = 47), full texts were not found (n = 31),  Men who have sex with men 3 5.9 People who use drugs 1 2.0 *where n ≥ 51 as some studies included multiple data types and/or participant groups.

Categories
Pooled rate The two-week morbidity rate 53% 95%CI:37-68% The two-week visit rate 20% 95%CI:24-67% The non-visit rate of PLWH with illnesses in two weeks 20% 95%CI:9-43% The annual inpatient rate 15% 95%CI:8-24% The rate of non-hospitalization among PLWH who required hospitalization 30% 95%CI:12-74% Fifty-one studies reported factors associated with healthcare access and utilization among PLWH. These factors were often interrelated and exerted their in uence at different levels in different contexts. This review identi ed the factors in each category based on the socio-ecological framework and illustrated them in Fig. 8.

Discussion
Summarizing the status of healthcare disparities among PLWH is essential for improving healthcare equity.
This review summarizes literature that examines disparities in healthcare access and utilization among PLWH in China and yielded 51 articles. It extends the existing literature by understanding the status of the disparity, assessing the driving forces of disparities from different levels, and providing evidence for future intervention.
Our review found substantial healthcare access and utilization disparities among PLWH in China. We also identi ed different factors affecting healthcare access and utilization.
In our review, disparities in receiving ART and ART adherence were found among PLWH in China. The rate of receiving ART among PLWH aged ≥ 45 years is higher than those aged < 45 years, which is different from a report from Canada [72]. Avoidant coping, fear of unintentional disclosure, and stigmatizing social norms may diminish the intention of young PLWH to initiate ART [73]. Similar to the South African study [74], we found that the rate of receiving ART is higher among women living with HIV than among men. HIV infection limits the ability to sexually behave, earn money and raise a family, and the resulting stigma undermines treatment seeking for men living with HIV [75]. In addition, people who are infected with HIV through injecting drug use appear to have a lower rate of receiving ART than people infected with HIV through sexual transmission, similar to the nding of a Spanish study [76]. Due to legal challenges, unstable housing, poverty, and lack of social and family support [77,78], people who inject drugs face multiple barriers to accessing health services, which largely delay ART initiation. The pooled rates of ≥ 95% ART adherence reported by PLWH in China (82%) is higher than that reported in India (77%) as found in a recent systematic review [79]. have accounted for our ndings. Additionally, the non-visit rate of PLWH with illnesses in two weeks and the rate of non-hospitalization among those who required hospitalization were both higher than the national general population (the non-visit rate of residents with illnesses in two weeks: 1.7%, the rate of nonhospitalization among residents who required hospitalization: 10.2%) [91]. This indicates that access to non-HIV-focused healthcare services in PLWH is inadequate.
Personal nancial constraints were the primary cited barrier between the intention to seek care and the actual use of care services. Although the nation-free ART program began in 2003, only PLWH with CD4 cell counts below 200 were eligible for free treatment [92]. The treatment criteria were updated in 2016 enabling all ART for all PLWH to access ART [93]. The strict eligibility criteria in earlier years led some PLWH to pay for ART out of pocket and deterred some from seeking care to date. Also, the costs of laboratory tests and expenditures related to adverse drug events (ADEs) are not covered or subsidized although the free ART policy covers the cost of antiviral drugs for PLWH. This may explain some of the nancial constraints cited by PLWH, as a Nigerian study showed that the incidence rate of ADEs among PLWH was high at 28.3% [94]. In addition, the reimbursement rate of medical insurance for the treatment of PLWH is low. For example, medical insurance only covers 12.7% of PLWH outpatient medical expenses in Dongguan City, while in Nanchang City, medical insurance covers 20.8% [95]. Expanded insurance policies, including coverage of more medical services and higher reimbursement rates, are needed to reduce the burden of medical costs on PLWH in China.
Fear of HIV status disclosure was also a common reason for the lack of care engagement and treatment retention. Many people diagnosed with HIV face di culties in deciding when and how to disclose their status to those around them. The anticipated negative consequences of HIV disclosure are considered an important driver in the disclosure process [96]. Some studies have shown that stigma is a major factor in non-disclosure [97,98]. Stigma is rooted in culture and driven by personal and social values [99]. HIV-related stigma stems from perceptions of "sexual immorality" and misconceptions about the mechanisms of HIV transmission [100].
In Chinese society, the phenomenon of prejudice against PLWH is widespread. A study among PLWH in central China showed a moderate to high level of perceived stigma [101]. More than 70% of healthcare workers had discriminated against PLWH in the delivery of medical services in healthcare settings [102]. Therefore, interventions to reduce HIV-related stigma such as enhancing social support, improving the quality of HIV care, and strengthening the capacity of healthcare providers to promote the availability of healthcare services among PLWH are necessary.
This review summarizes the status and the driving forces of healthcare access and utilization disparities among PLWH in China and provides directions for future interventions to promote healthcare equity. Creating a supportive social environment and expanding insurance policies could be considered to promote healthcare equity. However, there are some limitations to our scoping review. First, the included articles were published from 2000 to 2022, in which the free ART policy underwent several adjustments. Thus, part of our study before 2016 was based on the old policy, which may reduce the applicability of the results. Furthermore, the pooled rate of adherence estimated in this review may be overestimated as the 13 included studies used PLWH selfreports to estimate adherence. Another shortcoming is that studies included in this scoping review are only related to the Chinese context, which may limit generalizability. Nevertheless, the ndings of this scoping review will provide a good basis for further syntheses.

Conclusion
This review indicated great disparities in ART access, adherence, and utilization of non-HIV-focused care services among PLWH. Financial issues, social discrimination, structural problems in the health care system, and lack of psychological support are prominent in the process of utilizing medical services for PLWH.
Expanding current insurance policies, including covering more medical services and increasing reimbursement Page 15/30 rates are needed to ease the burden of medical costs for PLWH. Also, the government could negotiate with private insurance to subsidize the cost of covering non-HIV-focused services for PLWH. Creating a supportive social environment is needed to reduce HIV-related stigma and discrimination by using crowdsourcing to develop more innovative and participatory stigma reduction interventions.

Consent for publication
Not applicable.

Availability of data
All supporting data (and its additional les) is attached to this manuscript.

Competing interests
The authors declare that they have no competing interests.

Funding
This work was supported by the Key Technologies Research and Development Program (2022YFC2304900-4 to WT), National Institute of Health (R34MH119963 to WT), National Nature Science Foundation of China (81903371 to WT), and CRDF Global (G-202104-67775 to WT). The funders had no role in the study design, data collection, analysis, decision to publish, or preparation of the manuscript.

Authors' contributions
WT and GF conceived the study. WA and CF performed all database scans and articles for inclusion, generated all tables and gures, and drafted the rst draft. GF, JT, JO, DW, GM, and RT provided critical revision of the paper. All authors read and approved the nal manuscript. The distribution of study regions   Forest plot of PLWH receiving ART by ethnicity and transmission route. A1 presents a forest plot of receiving ART among PLHIV of Han ethnicity. A2 presents a forest plot of receiving ART among PLHIV of ethnic minorities. B1 shows a forest plot of receiving ART among people infected with HIV through heterosexual transmission. B2 shows a forest plot of receiving ART among people infected with HIV through homosexual transmission. B3 shows a forest plot of receiving ART among people infected with HIV through injecting drug use.

Figure 5
Page 28/30 Forest plot of ART adherence among PLWH.

Figure 6
Forest plot of ART adherence among PLWH by gender. A1 shows a forest plot of full ART adherence among men living with HIV. A2 shows a forest plot of full ART adherence among women living with HIV. B1 shows a forest plot of ≥95% ART adherence among men living with HIV. B2 shows a forest plot of ≥95% ART adherence for women living with HIV. C1 shows a forest plot of >90% ART adherence among men living with HIV. C2 shows a forest plot of >90% ART adherence among women living with HIV.

Figure 7
Forest plot of outpatient and inpatient service utilization among PLWH. A presents a forest plot of the two- week morbidity among PLWH. B presents a forest plot of the two-week visit among PLWH. C presents a forest plot of the non-visit rate of PLWH with illnesses in two weeks. D presents a forest plot of the annual inpatient among PLWH. E presents a forest plot of the non-hospitalization among PLWH who required hospitalization.