Changes in Access to Health Services during the COVID-19 Pandemic: A Scoping Review

The COVID-19 pandemic and the measures adopted are having a profound impact on a major goal of public healthcare systems: universal access to health services. The objective is to synthesize the available knowledge on access to health care for non-COVID-19 conditions and to identify knowledge gaps. A scoping review was conducted searching different databases (Medline, Google Scholar, etc.) for original articles published between December 2019 and September 2021. A total of 53 articles were selected and analyzed using the Aday and Andersen framework as a guide. Of these, 37 analyzed changes in levels of use of health services, 15 focused on the influencing factors and barriers to access, and 1 studied both aspects. Most focused on specific diseases and the early stages of the pandemic, based on a review of records. Analyses of the impact on primary care services’ use, unmet needs or inequalities in access were scarce. A generalized reduction in the use of health services was described. The most frequent access barrier described for non-COVID-19 conditions related to the services was a lack of resources, while barriers related to the population were predisposing (fear of contagion, stigma, or anticipating barriers) and enabling characteristics (worse socioeconomic status and an increase in technological barriers). In conclusion, our results show a general reduction in services’ use in the early stages of the pandemic, as well as new barriers to access and the exacerbation of existing ones. In view of these results, more studies are required on the subsequent stages of the pandemic, to shed more light on the factors that have influenced access and the pandemic’s impact on equity of access.


Introduction
The pandemic due to coronavirus SARS-CoV-2 (COVID-19), a novel virus initially reported in December 2019 [1], was declared by the World Health Organization (WHO) on 11 March 2020. It has had an adverse effect worldwide on many different spheres of society, including the economy and public health. The current COVID-19 pandemic and the diverse strategies that have been adopted to tackle it are forcing changes in access to health services for other conditions, potentially producing an impact on the health of the population above and beyond that caused by COVID-19 itself [2][3][4][5][6].
In this regard, some strategies taken to combat soaring COVID-19 infection rates may have negatively affected access to health services for other conditions. Firstly, at the health services level, one of the most influential measures was the classification of services as essential or non-essential, following WHO guidelines, which allowed resources to be redirected to the pandemic response. However, this has also caused cancellations or delays in elective and non-urgent procedures [3,[5][6][7], despite many countries implementing strategies to mitigate the impact of these disruptions (e.g., online healthcare visits) [4,5], Another significant measure was social distancing to reduce interaction between people, including nationwide partial or complete lockdowns, schools and non-essential business closures, and instructions to stay at home, which in some cases has erected a barrier in terms of mobility to entry the health services [8].
Another indirect effect of the pandemic, the economic crisis stemming from the substantial curtailment of economic activity, and the ensuing rise in unemployment and loss of household income, have aggravated associated access barriers (loss of health coverage, difficulties in making copayments or obtaining transport to services), thereby accentuating existing inequalities in access, as studies on previous economic crises have shown [9]. Although it is necessary to evaluate which population groups have been particularly affected in terms of access to care and how the determining factors interact with each other, there are some signs-including early evidence and experiences from previous crises-to indicate that vulnerable population groups (populations with low socioeconomic status, the elderly, chronic patients or those with severe conditions, migrants from low-income countries) suffer a greater impact [9][10][11][12][13][14].
In short, as in other epidemics and previous outbreaks, the health repercussions of the current pandemic are not confined solely to COVID-19 infection and mortality. They also include indirect negative effects on healthcare access and on the quality of curative and preventive care provided for other conditions, and the exacerbation of difficulties and barriers related to socioeconomic factors [15][16][17]. The scientific evidence accumulated from previous experiences, such as severe acute respiratory syndrome or SARS (2002)(2003), Middle East respiratory syndrome or MERS (2012), Ebola (2014-2016; 2018-present), and the Zika virus (2015-2016) [18][19][20], shows a decrease in the utilization of health services (e.g. outpatient care, hospital admissions, elective surgeries [20][21][22][23]) that is attributed to changes made to the health services in response to public health emergencies, as well as to fear of contagion among the population [21,24]. This may in turn have had an impact on increasing the burden of disease and mortality in the months following an epidemic outbreak [6,20,[25][26][27].
While a plethora of scientific papers have been published on COVID-19 since the start of the pandemic, studies on its impact on access to health services have not been so plentiful. A few literature reviews have been found that summarize changes in health services due to the pandemic, focusing mainly on the adoption of telemedicine [28][29][30][31][32] and the impact of the pandemic on different aspects, among others, the use of certain specific services, such as maternal and child health care [33], child vaccination [34], or chronic diseases [35,36], in the initial stages of the pandemic. Although we are still at the pandemic management stage, a synthesis of the scientific evidence accumulated to date on the impact on access to health care in different contexts, in addition to detailed monitoring of the performance of services, may help decision makers to make healthcare systems more resilient in current and future emergencies and protect public health and access to health care.
Access to care involves many highly interdependent factors and stakeholders at play [37]. This study takes as its framework of reference one of the most frequently used models for the analysis of access to health services [38], that of Aday and Andersen [39]. This model distinguishes between realized access (effective utilization of the services) and potential access (determinants of access), differentiating between factors related to the services and to the population. The realized access analysis takes into account the type, place, motive of the visit (preventive or curative), and care outcomes, while potential access analysis takes into account the characteristics of the services (availability of resources and organization) and those of the population (predisposing factors: sociodemographic factors, beliefs, attitudes, and knowledge of the health system; enabling factors: income and type of insurance; health needs). Health policies, programs, or interventions can in turn affect access barriers related to the health services or changing (mutable) characteristics of the population [39,40]. Aday and Andersen's framework is more comprehensive and exhaustive than other analytical models [41][42][43], which focus either on the entry to health services or on the characteristics of services and how they adjust to the characteristics of the population. Thus, it offers an appropriate approach for identifying existing knowledge gaps in the literature on access and to analyzing different types of barriers and factors that influence the use of health services.
The aim of this article is to synthesize the knowledge accrued from the onset of the pandemic in March 2020 through to September 2021 on the impact of the COVID-19 pandemic on access to health services (including primary care, outpatient secondary care, and inpatient care) for non-COVID-related conditions, and to identify knowledge gaps on these subjects.

Materials and Methods
A scoping review of the scientific literature [44] was carried out, following the PRISMA guide [45], to identify original articles on the impact of the COVID-19 pandemic on access to health services for non-COVID-related conditions.
In our bibliographic search, several digital databases were consulted to minimize the risk of overlooking any relevant studies: Medline, Google Scholar, SCiELO, and Web of Science. The search was performed over two separate periods: 22 January 2021-31 March 2021 and 22 September 2021-10 October 2021. In the Medline database, using a thesaurus, MeSH terms were employed for: (a) COVID-19: "Coronavirus Infections", "Coronavirus", "COVID-19", "SARS-CoV-2"; (b) access to health services: "Health services availability", "Health services needs and demand", "Healthcare disparities", "Needs assessment, healthcare", "Health policy", "Equipment and Supplies Utilization", "Facilities and Services Utilization" (see Appendix A for more detail). In the other databases free text terms were used: (a) COVID-19: "coronavirus disease", "COVID-19"; (b) access to health services: "health services accessibility", "accessibility", "accessing", "access", "utilization", "delivery of health care", "healthcare services", among others. Word groups were combined using Boolean operators "AND" and "OR" in order to identify the literature in the databases consulted and select those studies referring to the impact of COVID-19 on access to health services. The search was complemented with a manual review of references cited within the selected articles.
We selected original articles published in Spanish or English from December 2019 to September 2021-with no filters for geographical area or motive for access (apart from being unrelated to COVID-19)-that used qualitative and/or quantitative methods and analyzed or described changes in access to health services in the context of the COVID-19 pandemic. The initial selection of studies to review was performed through title and abstract screening. Where there was any doubt about whether to include a study, this was discussed with another researcher in the team.
Following Aday and Andersen's framework [39,40], the selected studies were classified into two groups: those that analyzed utilization of services (realized access) and those that explored factors that influence access (potential access). Any studies on realized access that did not use medical records, administrative/institutional databases, or patient surveys as their data source were excluded. A data extraction protocol was produced to include information related to methodological aspects (methods, period of analysis, study area, population, sample, type of health service) and study results (according to the variables or dimensions of analysis). This information was extracted from the articles and presented in tables, grouped according to type of access, and ordered by type of health service. The results were summarized according to the analytical framework, which was also used to identify any gaps in knowledge related to the aspects studied.

Results
From the search results, 242 articles were identified for title and abstract screening, and 95 for full-text review. A total of 53 articles met the inclusion criteria for analysis ( Figure 1). Of the 53 articles selected, 37 analyzed changes in realized access applying quantitative methods , through the analysis of medical records (Table 1), 8 analyzed changes in potential access [83][84][85][86][87][88][89][90] via surveys of different population groups, except for one that did the same via analysis of medical records [90] (Table 2), 7 used qualitative methods to analyze the impact on potential access [91][92][93][94][95][96][97] (Table 3), and, finally, 1 study analyzed changes in both realized and potential access [98] using mixed methods (analysis of medical records and semi-structured interviews). Only six of the studies that explored changes in potential access included professionals from the health centers analyzed as a study population, [87,89,92,93,97,98], while the rest focused on patients or the general population. General Reduction in the number of childcare visits (50%), including preventive procedures, and sustained utilization of HIV services and adult outpatient clinics during the national lockdown (March-June 2020), compared to the previous periods. No significant changes were observed at a general level in the use of services. Childcare visits recovered over the following three months to pre-lockdown levels.  Reduction in visits to pediatric emergencies (57 to 70%), especially during the peak of the pandemic (March-April 2020), compared to previous periods. Admission proportion almost doubled (4% pre-pandemic to 7% during the peak pandemic period). Average acuity of illness was higher during the pandemic period. Reduction in the monthly number of referrals for suspected cancer (63%) and for treatment (22%), colonoscopies (92%), and surgeries (31%) from April 2020, compared to 2019 and the preceding months. Relative increase in radiotherapy use (44%) due to increased use of short-course regimens.
Below is a summary of results found regarding changes in realized access and potential access, following the Aday and Andersen theoretical framework [39,40].

Changes in the Utilization of Health Services and Influencing Factors
Of the 38 studies that analyzed changes in realized access 98], 33 indicated a statistically significant descent in the use of services and only one reported an increase [78], and 4 descriptive studies also found a reduction in the use of services [47,61,63,65] (Table 1). After the first few months of the pandemic, some studies described an increase in the utilization, without reaching levels previous to the COVID-19 pandemic [48,61,63,71,72,82]. However, among the studies that extended their analysis to the end of 2020, there are reports of new drops in the utilization of services, coinciding with the onset of new waves of COVID-19 [61,63,71].
By type of service analyzed, studies focusing on the health services in general [46,47,49,50] described an overall drop in use, which varied in terms of magnitude. Among those that analyzed changes in emergency care [51][52][53][54][55][56][57][58][59][60][61][62]66,77], it was reported that, although the volume of consultations fell, there was an increase in the number of cases with additional complications [53,58,66,74,77] and cases requiring admission to hospital [51,55,57,59,60,67,70]. Likewise, the studies on changes in SC services also reported a reduction in the use of services, while in some cases detecting an increase in the proportion of visits or hospital admissions in concrete healthcare areas (e.g., severe mental health cases [47] or trauma injuries [56,58,78]). With respect to PC, both studies described a drop in the number of in-person visits and an increase in remote care consultations (via various mechanisms, such as consultation by telephone or videocall) [66,80]. Lastly, the study that analyzed changes in the use of preventive services [82], along with others that also assessed procedures classified as elective [49,53,63,64,77], reported a drop in use but with less pronounced changes in urgent cases [57,62,82].
In the studies that analyzed influencing factors in the use of services in the context of the pandemic, the probability of lower utilization levels was associated with different factors. With regard to predisposing characteristics of the population, women [56,79] and ethnic minorities [49,74] were less likely to access health services, with inconsistent results regarding the elderly [52,79]. As for individual enabling characteristics, the likelihood of a lower use of services was associated with people with a low income and limited healthcare coverage [49,74], and for those enabling characteristics related to the type of area, a higher incidence of COVID-19 [50]. In terms of need, the probability of using health services was lower among patients who lacked a previous diagnosis [82], had less severe conditions (without complications or adverse outcomes) [67], and did not require hospitalization [52,57,67].
Fourteen studies identified barriers related to population characteristics [83][84][85][86][87][88][89]91,[93][94][95][96][97][98]. Among the predisposing factors identified, fear of contagion was one of the main reasons for not going to the health services [83][84][85][86]88,89,[94][95][96][97][98], as well as the stigma that receiving a COVID-19 diagnosis would create [89,94,97]. Other factors found included people misinterpreting government recommendations to avoid going to the health services [84], perceiving that the services were of poor quality [94], or believing they would have difficulties in gaining access [85,86,91,93] during the pandemic. With respect to enabling factors, authors highlighted the worsening socioeconomic situation of the population [87,89,91,93,94,96,97], a lack of support networks [93,94], and an increase in technological barriers [85,88] as some of the main factors that hindered access to the health services. Lastly, they also underlined tendencies to play down the risk of health complications and the perceived need for medical attention [94,98] as barriers that had the effect of reducing the use of services and delaying the decision to seek care during the pandemic.

Discussion
The impact of the COVID-19 pandemic has been felt worldwide in many different spheres of society, but especially in access to health services for unrelated conditions. There is now a pressing need to evaluate the changes that have arisen in this regard, and their implications for equity of access and the resilience of national health systems to future pandemics. This is, to the best of our knowledge, the first scoping review to offer a general overview of the subject, taking in the current evidence and highlighting the areas that will require further research in future studies.
Most of the studies included in the analysis describe a lower level of health services' utilization and changes in potential access, as preexisting barriers have intensified and new ones have arisen. However, while investigations into the impact of the COVID-19 pandemic are still ongoing, the results of this review show that the studies conducted to date are limited in terms of scope and methodology, and that they are mainly centered on analyzing the impact on the use of services for specific diseases or population groups during the first stages of the pandemic, with a particular focus on secondary care.
Studies on the use of health services in general are very scarce, as are those on access to primary care, which in many countries has been the most overwhelmed care level due to having to take on more pandemic-related care duties (vaccination, case tracking, etc.). Likewise, there is a considerable lack of evidence so far on how changes have taken place over the course of the different waves of COVID-19, and according to geographical context. Although some studies with longer timeframes-to the end of 2020-have already described new drops in health services' activity following brief periods of recovery [61,63,71], further evidence is required to confirm this trend. Moreover, we have yet to look into how the pandemic has affected unmet care needs, as some studies showed that one of the most commonly reported impact was that patients delayed seeking care due to factors such as fear of contagion, disinformation, etc. [83][84][85][86]88,89,94,95,97,98].
It is important to bear in mind that access to care involves multiple interdependent factors and numerous actors. However, the results of this review show that, so far, there are almost no published studies with a wide scope using mixed methods, and that the qualitative studies available to date are still limited in both number and perspective. In regard to the latter, few studies include, in addition to that of users, other viewpoints such as that of the health professionals or managers involved in the process of taking measures or adopting practices that influenced access to care. Furthermore, the population groups selected for study were generally sufferers of a specific condition or of vulnerable status. The number of qualitative studies is probably limited due to the complexity of their development in the pandemic context, in terms of time, resources, and restrictions imposed by the mitigation measures. Approaches that combine multiple sources of evidence and different perspectives are needed to shed more light on the factors and actors that have influenced access.
With regard to our main findings on the reduction in the use of services, this may be related to health systems prioritizing their response to the public health emergency, which differed according to context [4,25,99,100]. Initial measures were generally centered on containing the spread of COVID-19 and providing the health services with the resources needed to meet the soaring demand for medical attention, which led to the classification of some services and procedures as non-essential, and a consequent reduction in the resources allocated to cover those health needs [3,4,7,8]. There is a lack of evidence on how the measures were modified and adapted according to context as the pandemic progressed, and the patterns of utilization that they generated, although some studies have already revealed changes in use with recoveries and relapses, in the more advanced stages of the pandemic [61,63,71].
Going into more detail, the reductions in access to services described appear to have brought with them an increase in medical complications and emergencies, especially in elective procedures [48,49,52,53,58,59,67,71,74,77,101], and/or care delays (time passed between onset of symptoms and intervention) [53,98], although it is generally not specified whether the delays were due to patients putting off seeking medical attention or rather to an increase in barriers to access the services. Some studies also observed higher mortality rates [52,70,74,76] and burden of disease [62]. However, while emergency care was one of the greatest causes for concern [51][52][53][54][55][56][57][58][59][60][61][62], according to our results the downturns recorded for this type of service appear to be less pronounced or of lower impact than those reported in other fields.
In this regard, some studies highlight the difficulties involved in maintaining normal levels of activity in certain services, even in some classed as essential, such as maternal health, oncology, or mental health [71,76,85,89,93,94,97,98]. The impact appears to have been felt worldwide but especially in middle-and low-income countries, a point that has also been stressed in various opinion articles [101][102][103][104][105][106][107]. Differences between health systems, and between geographical contexts, may both have acted as determining factors in the changes seen. Sexual and reproductive health services analyzed in African and South Asian countries, for example, have seen a significant downturn in access, not only to maternal and child health services but also other non-essential areas of health care (family planning, sexually transmitted diseases, etc.) [71,74,76,93,98], despite the warnings given in various reports and opinion articles based on previous epidemic experience [2,26,101,108] of the risk this poses in terms of burden of disease and mortality [2,25,26,[103][104][105][106][109][110][111][112][113][114]. Another example can be found in the lack of care for patients at risk of a cancer diagnosis due to delays in screening and diagnostic programs caused by certain procedures being classified as elective [64,65,82], which could also lead to an increase in the burden of disease and mortality due to the late detection of new cancer cases, as several studies pointed out [35,[115][116][117][118]. As regards mental health, various studies mention anxiety and other disorders related to fear of contagion and the uncertainty that the pandemic generated in its first few months [35,55,77,[83][84][85][86]88,91,[94][95][96], as well as difficulties in access to mental health services [68,85,97] and an increase in acute cases of severe disorders [67]. These issues may have caused the burden of disease to increase, which could influence service utilization patterns in the years to follow, an aspect that was observed previously with the SARS epidemic in 2003 [119] and that should be taken into consideration in future studies on the impact of COVID- 19.
The few studies that analyze the influencing factors on the lower utilization of services mainly highlight a greater downturn in use among low-income users, those with limited healthcare coverage, and ethnic minorities [49,74], as well as the female population [56,79], all of which signals greater inequalities with regard to more vulnerable populations.
As regards the changes in potential access detected in this review, the results indicate both the exacerbation of existing barriers, related above all to structural difficulties and situations of vulnerability, and the creation of new ones. In terms of existing barriers accentuated by the pandemic, some studies reported a shortage of resources in the services to meet all the incoming healthcare requests [83,85,86,88,89,92,93,95,97,98], which varied according to service and geographical context. One of the most serious problems was the lack of or alterations to the distribution of materials and medical supplies in lowincome countries [83,87,89,93,94,97], an aspect corroborated in other publications, both reports and opinion articles [4,34,104,[120][121][122]. The lack of materials and medical supplies may have contributed to increasing negative perceptions of the quality of the health services, especially in disadvantaged settings or situations with structural difficulties, another barrier to access found by some studies [93,94,97]. Lastly, several studies in this review [83,85,87,89,91,93,97] and some reports [12,123] focusing on vulnerable population groups (such as migrants or refugees, sex workers, people at a severe socioeconomic disadvantage) highlight the worsening economic situation and the intensification of other barriers (legal, information-related, or discriminatory), as has also occurred in previous epidemics [19]. All this points to situations in which the ability of these groups to access the health services and receive care may have diminished yet further.
In addition, new barriers may have been created as a result of adopting alternatives to face-to-face visits (online consultation, telephone, video call, telemedicine, etc.) and changes in attitudes and beliefs developed as a result of the pandemic. In this regard, the use of online consultation has grown as a way to mitigate difficulties in access [4], but not in the same way across all contexts [3,124]. Several articles included in the review described inequalities in access, reporting access problems related to digital literacy (lack of understanding of digital devices) or a lack of material resources (Internet connection, mobile devices) [49,85,89], a point also highlighted in other publications [125,126], alongside the perception that the care received in virtual consultation is impersonal [88]. These results are in keeping with numerous publications that have focused on the changes from face-to-face consultations to remote care and user satisfaction with the latter [29,80,[127][128][129][130][131][132][133][134][135][136]. However, most studies are centered mainly on high-income countries, thus, further evidence in different contexts is required on the impact of remote care on access to the health services.
While it is true that various studies have identified both new barriers and the exacerbation of existing ones, the behavior of individuals in this type of public health emergency requires more in-depth analysis in order to steer the design of interventions to help counter these barriers, such as public health information campaigns or specific measures for vulnerable populations.
On a final note, this review has several limitations. In the first place the nature of the studies covered varies greatly, in terms of methodology (ways of measuring use of services, information sources, sample size, etc.), and of geographical areas and health systems analyzed; thus, it is not possible to draw comparisons between them. Secondly, articles that analyzed access to services due to COVID-19 in addition to other illnesses were excluded from the study. This decision was made in order to rule out bias towards activity and resources destined to the treatment of other diseases. Third, no studies focusing on the impact of the pandemic control measures on access to health services were found, probably due to the limited terms used in our search to capture this area. Moreover, it is also possibly due to the difficulties involved in distinguishing the impact of the measures from other effects of the pandemic. Finally, some articles may not have been considered on being published in other languages (Chinese, Arabic, etc.), so this analysis may have excluded relevant information and failed to consider certain contexts. In spite of these limitations, this is the first study to address changes in access from a global perspective, with a view to shedding light on gaps in knowledge that will require further research in the future.

Conclusions
This review analyzed studies that reported changes in health services' utilization, and the factors that influenced the use of services for non-COVID-19 conditions, during the COVID-19 pandemic. Results vary according to the context analyzed, although, in general terms, they reflect the same trend, describing a general reduction in the use of health services, the exacerbation of preexisting barriers, and the emergence of new ones. This scoping review has shown that most studies conducted to date are limited in terms of scope and methodology and are centered mainly on the impact on specific conditions or population groups during the early stages of the pandemic, focusing mostly on secondary care. Furthermore, a significant gap in knowledge was detected on whether the services have recovered to pre-pandemic levels of care use and, if not, in which areas and for what reasons. Future studies should go into greater depth on the pandemic-related changes that have influenced access to health services (e.g., fear and socioeconomic difficulties), according to context and over the course of the different stages of the pandemic. In any case, as an ongoing phenomenon, the real impact of the COVID-19 pandemic is yet to be determined. Funding: This research received no external funding. G.P. received the CSC grant to conduct this study as her thesis for the Master in Public Health (Universitat Pompeu Fabra).

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Conflicts of Interest:
The authors declare no conflict of interest.