key: cord-0862375-ka0z6nsg authors: Han, Bingxue; Guan, Hongyi title: Associations between new health conditions and healthcare service utilizations among older adults in the United Kingdom: effects of COVID-19 risks, worse financial situation, and lowered income date: 2022-04-22 journal: BMC Geriatr DOI: 10.1186/s12877-022-02995-8 sha: 784e7181c4400f3a8ada75dc305e2d43e9f55526 doc_id: 862375 cord_uid: ka0z6nsg BACKGROUND: Health services are critically important for older adults, particularly during the Coronavirus disease-19 (COVID-19) pandemic. However, COVID-19 risks, worse financial situation, and lowered income may seriously impact health services by feasibility and accessibility. Therefore, the aim of the present study was empirically to explore how health-seeking behaviors are influenced by new health conditions through COVID-19 risks, worse financial situation, and lowered income. METHODS: Data were from ELSA COVID-19 waves 1 and 2 which included a sample of 6952 and 6710 older adults in the United Kingdom, respectively. The frequency distribution analyses were conducted by Chi-square analysis by gender groups. Zero-inflated Poisson regressions were used to examine how worse financial situation and lowered income were associated with COVID-19 risks and new health conditions. Logistic regressions were employed to examine the associations of COVID-19 risks, worse financial situation, and lowered income with treatment cancellation and accessible care. Cross-sectional mediation models, cross-sectional moderation models, longitudinal mediation models, and longitudinal moderation models were conducted based on Hayes model 6, Hayes model 29, Montoya model 1, and Montoya model 2, respectively. RESULTS: Most of the sample was >65 years old, females, located in urban place, and involved in long-standing condition. Regression analysis showed that COVID-19 risks, worse financial situation, and lowered income were associated with treatment cancellation and accessible care. In the longitudinal mediations, effect coefficients of ‘X’ → (treatment cancellation in wave 1 (Tcn1)- treatment cancellation in wave 2 (Tcn2))(β = −.0451, p < .0001, low limit confidence interval (LLCI) = −.0618, upper limit confidence interval (ULCI) = −.0284), ‘X’ → (COVID-19 risks in wave 1 (Csk1)- COVID-19 risks in wave 2 (Csk2)) (β = .0592, p < .0001, LLCI = .0361, ULCI = .0824), and ‘X’ → (lowered income in wave 1 (CIn1)- lowered income in wave 2 (CIn2)) (β = −.0351, p = .0001, LLCI = -.0523, ULCI = -.0179) were significant. Additionally, effect coefficients of ‘X’ → (accessible care in wave 1 (Acr1)- accessible care in wave 2 (Acr2)) (β = .3687, p < .0001, LLCI = .3350, ULCI = .4025),'X’ → (Csk1- Csk2) (β = .0676, p = .0005, LLCI = .0294, ULCI = .1058), and ‘X’ → (worse financial situation in wave 1- worse financial situation in wave 2) (β = −.0369, p = .0102, LLCI = -.0650, ULCI = -.0087) were significant. CONCLUSIONS: There were longitudinal mediating effects of COVID-19 risks, worse financial situation, and lowered income on the relationship between new health conditions and treatment cancellation and relationship between new health conditions and accessible care. These findings suggest that worse financial situation, lowered income, and COVID-19 risks exerted an influence on the relationship between new health conditions and treatment cancellation and relationship between new health conditions and accessible care among older adults. Findings suggest that longitudinal mediations may be important components of interventions aiming to meet service needs. Long-term health policy implications indicate the need for reducing COVID-19 risks, improving financial situation, and increasing income among the targeted population. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12877-022-02995-8. Selected health conditions often professionally correspond to likelihood of improvement with treatment and care. But, the prevalence of co-occurring psychiatric disorders, co-existing chronic health conditions, and concurrent behavioral health conditions often was high when diagnosed a specific disease [1, 2] . Thus, treatment challenges [3] , treatment utilization [4] , and treatment needs [5] , and treatment effectiveness [6] were reported in such settings. Subsequently, patients' perception regarding medication effectiveness could strongly lead to treatment satisfaction [7] and barriers for treatmentseeking [8] . Objectively, self-reports of treatment for secondary health conditions [9] , affordable care [10] , and mental health problems [11] were identified as potential predictors of treatment. Moreover, co-occurring health conditions often need collaborative care in the treatment [12] and integrating treatment [13] . Empirically, the association between health conditions demonstrates the need to consider in the context of other health hazards or different healthcare settings [14] . Early studies documented health-seeking behaviors among the persons with poor health conditions. For example, a cross-sectional survey in Albania concluded public facilities were reported as the main health care service providers by adults with non-communicable diseases [15] . An Australian cohort demonstrated that persons with hypertension and poor self-rated health were likely to visit frequent monthly general practitioners [16] . An observational study indicated improved healthcare access led to improved physical and mental health [17] . Thus, persons with health conditions may behave as health-seeking behaviours. Factors including stigma [18] , treatment costs [19] , patient views [20] , and health consultant [21] possibly lead to under-treatment of health conditions. In addition, socioeconomic disparities were identified in the reporting of health conditions and the use of discretionary health services [22] . In addition, a study indicated that uninsured individuals in poor health conditions consumed less health care [23] . Dynamic treatment regimes [24] , personalized care [25] , and accessible personal medicine [26] are considered as ideological and effective methods to manage and improve health status. Patients with acute health conditions were more likely to require hospitalization than those without. Collaborative care models [27] and interdisciplinary multi-professional treatment [28] are valuable approaches for patients with psychiatric and physical health conditions in reallife hospital settings. It was confirmed that web-based self-care interventions for chronic conditions could not be replaced with outpatient and inpatient treatment to reduce health inequalities [29] . Regarding early impacts of the COVID-19 pandemic on mental disorders, a review found deteriorations in symptoms, lack of access to services and resources, and self-management in inpatient and residential settings [30] . It was confirmed that the fear of COVID-19 could trigger preventive health behaviors [31] . In the time of COVID-19, persons with outdoor activities appear to be more susceptible to the coronavirus disease than before. During the Coronavirus disease-19 (COVID-19) period, a lot of studies reported change of health-seeking behaviors caused by COVID-19. Multiple studies identify the healthcare and emotional needs [32] , use of new technology for social connectedness [33] , accessibility and utilisation of telehealth [34] among older adults during the COVID-19 pandemic. Empirical studies have reported self-cancelled appointments [35] , health-check cancellation [36] , cancelled surgeries [37] , canceled appointments [38] , treatment pause [39] , therapy postponements [40, 41] , fertility treatment cancellation [42] due to COVID-19 pandemic. Delay in fertility treatment [43] , fertility treatment suspensions [44] , and cancelled appointments [45] were also reported during COVID-19 pandemic. Thus, COVID-19 pandemic may significantly influence utilizations of health care service of older adults. In this context, the goal of this study was to explore the mediating factors with ELSA COVID-19 with data collected in representative samples of UK, in 2021. The main hypothesis in this study was that the relationships of new health conditions → treatment/care would be mediated/ moderated by barriers to accessing health services. Here, barriers to accessing health services included COVID-19 risks, worse financial situation, and lowered income. The main relationships of interest were motivated on the basis of Health Belief Model (HBM). HBM empowers researchers to explain and predict health promoting behaviour in terms of patterns of belief by addressing the association between personal threat of an illness or disease and health services utilisation [46] [47] [48] . Facing the susceptible risk of COVID-19 pandemic, the public population with new health conditions would like to perform care/treatment-seeking behaviors when considering the related severity and barriers. In this study, mediators /moderators included COVID-19 risks, worse financial situation, and lowered income. A large body of research documented economic burden and needs of health care service. For example, a study indicated that COVID-19 lockdown reduced numbers of patients living in poor socio-economic conditions accessing the mental health service [49] . A study among rural migrant workers in China indicated that unequal health service utilization sourced from income-related inequality [50] . A study in Japan concluded change in household income might influence the utilization of long-term home care services [51] . Simultaneously, multiple studies documented financial barriers to access to health services [52] [53] [54] [55] . Forgoing necessary medical treatments [56] and refraining from health care [57] were reported. Financial hardship and health risk behavior were reported during COVID-19 in a large US national sample of women [58] . A study in among African American men suggested that financial hardship can lead to unmet medical need due to cost [59] . A study in central Malawi indicated financial accessibility of health services was identified the determinants of health seeking behaviours [60] . Several studies indicated part of patients did not seek further treatment testing due to low socio-economic status [61, 62] . Socioeconomic factors regarding early signs and symptoms resulted in delayed diagnosis [63] . Furthermore, the general population is experiencing more new health conditions than ever given the current COVID-19 pandemic. COVID-19 has been confirmed as a significant impact on the quality of life, general health, and social relationships, and illness acceptance of patients [64] . Here, this study employed publicly survey data to investigate the roles of COVID-19 risks, worse financial situation, and lowered income in the associations between new health conditions and healthcare service utilizations. Academically, mediation analysis is applied to quantify the causal sequence "predictor → mediators → outcome" for observational studies within path models. Moderation describes a situation in which the relationship "predictor → outcome" depends on the values of moderators. Since COVID-19 pandemic, multiple studies documented mediating/moderating roles of care/ treatment in the relationship of interest. For example, a study treated healthcare service utilization as a mediating variable [65] . Another investigation treated new health condition as a mediating/moderating variable [66, 67] . Analogously, outcome variable in a similar research was consequent mental health status rather than healthcare service utilizations [68] . As compared to cross-sectional mediation/moderation, longitudinal mediation/moderation could reflect changes in mediating/moderating effects over time and lead to more accurate representations of the temporal precedence of change over time. Here, the conceptual frameworks of potential crosssectional and longitudinal mediation/moderation are designed in Figs. 1, 2, 3, 4, 5, 6, 7 and 8. In this study, we hypothesize that COVID-19 risks, worse financial situation, and lowered income mediate/ moderate the relationship between new health conditions and healthcare service utilizations. To confirm the hypotheses of interest, the present study would quantify mediators and moderators with cross-sectional and longitudinal mediation/moderation models. The hypotheses of this study are as follows: H1a: The associations of new health conditions with treatment cancellation would be mediated by COVID-19 risks, worse financial situation, and lowered income in cross-sectional mediation models. H1b: The associations of new health conditions with accessible care would be mediated by COVID-19 risks, worse financial situation, and lowered income in cross-sectional mediation models. H2a: Worse financial situation and lowered income moderated the effect of COVID-19 risks on the associations of new health conditions with treatment cancellation in cross-sectional moderation models. H2b: Worse financial situation and lowered income moderated the effect of COVID-19 risks on the associations of new health conditions with accessible care in cross-sectional moderation models. The empirical findings of this work would contribute to the related academic fields. Regarding theoretical implications, our study would extend prior research on the HBM application for the relationship between new health conditions and healthcare service utilizations with crosssectional/longitudinal mediation/moderation. Considering practical implications, our study would provide This study employed a survey-based data from The English Longitudinal Study of Ageing (ELSA) COVID-19 Substudy (https:// www. elsa-proje ct. ac. uk/ covid-19). The ELSA COVID-19 Substudy allows a cross-sectional analysis of the dynamics of the lockdown, health changes, economic vulnerabilities, and health-seeking behaviors. This study has been reviewed and approved by the University College London Research Ethics Committee. The ELSA sample is based on respondents aged 50 and over who participated in the Health Survey for England. The respondents from three years of 1998, 1999 and 2001 consisted of the original sample and provide a sufficiently large sample size (23,132 responding households). In the course of the study, the sample has refreshed at the younger age range to maintain the 50+ design. The ELSA COVID-19 survey questionnaire covers the following topic areas: demographics, mental health, financial security, COVID-19-related health, employment and work, financial situation, volunteering and care, physical health and health behaviours, social connection isolation and technological inclusion, and income, pensions and retirement. The first and second wave of the substudy lasted 54 days (from the 3rd June 2020 to the 26th July 2020) and 47 days (from the 4th November 2020 to the 20th December 2020), respectively. The first wave of the ELSA COVID-19 Substudy aimed to understand the immediate impact of the COVID-19 crisis on health, access to health and social care, financial circumstances, mental wellbeing, and social activity in the older population in England. It also aimed to set a baseline for a second wave, which looked to assess what changes had taken place in mental and physical health, finances, and social experience of the older population by the end of 2020. As for the ELSA COVID-19 Substudy, both cross-sectional and longitudinal weights were developed to maximize the available sample size and increase the statistical power of the analysis in conjunction with ELSA Wave 9. ELSA Wave 9 received ethical approval from the South Central -Berkshire Research Ethics Committee on 10th May 2018 (17/SC/0588). The cross-sectional weights in COVID-19 Substudy wave 1 included weight for core members only and for both core members and partners. The longitudinal weights in COVID-19 Substudy wave 2 were established for analyses of wave 1, wave 2, ELSA wave 9.Furthermore, given the relatively high response rate from partners, an additional weight was created to allow partners aged 52+ to be included in weighted analysis for COVID-19 Substudy waves 1 and 2. Considering COVID-19 Substudy sampling approach, all participants for the COVID-19 substudy were selected from the existing ELSA sample. Study participants issued to wave 1 who were still eligible ahead of wave 2 and did not request to leave the COVID-19 Substudy or the ELSA study were issued to wave 2. Number of issued study participants was 9525 in wave 1 and 9150 in wave 2, respectively, while number of issued core members was 7689 in wave 1 and 7465 in wave 2, respectively. Number of completed survey interviews were 7040 in wave 1 and 6794 in wave 2, while number of productive core members were 5825 in wave 1 and 5338 in wave 2.With a sequential mixed-mode strategy (web + telephone) data collection, the ELSA COVID-19 Substudy achieved a final response rate of 75% (online: 83% and phone:17%). ELSA is a so ripe and famous survey that it can guarantee complete confidentiality and avoid identification of individuals or organisations. All methods were carried out in accordance with relevant guidelines and regulations. Written informed consent was obtained from all participants before they agreed to participate in the study. Participants were informed that they could leave the study at any time without penalty, and all personal information was kept confidential. Healthcare service included treatment cancellation and accessible care. Treatment cancellation was reflected by the question: "Since the coronavirus outbreak, have you had a hospital operation or treatment cancelled?" with responses of no and yes in wave 1 and no, yes and "I did not have a hospital operation or treatment booked" in wave 2. The answer "I did not have a hospital operation or treatment booked" was treated as a missing value in wave 2. Thus, the response options were no (=0) and yes (=1). Accessible care were reflected by the question: "Since the coronavirus outbreak, have you been able to access the community health and social care services and support you need, for instance a dentist, podiatrist, nurse, counselling or personal care?" with responses of no, yes, "I did not attempt to contact them", and "I did not need to contact them" in wave 1 and no, yes, and "I did not need to contact them" in wave 2. The answers "I did not attempt to contact them" and "I did not need to contact them" were treated as a missing value in waves 1 and 2. Thus, the response options were no (=0) and yes (=1). Infected risk for COVID-19 consist of five questions: "Has anyone in your household tested positive for COVID-19?" , "Has anyone in your household had to stay in hospital for treatment due to COVID-19?" , "Has anyone close to you outside your household tested positive for COVID-19, for example a relative or a friend?", "Has anyone close to you, such as a friend or family member, died with COVID-19?" , and "Has anyone in your household died with COVID-19?" Their response options were yes (=1) and no (=2). For convenience, missing values of the response options in the five questions are defined as zeros. Here, the response options were recoded as binary response (yes = 1, no = 0). Then, sum response options could be calculated and obtained COVID-19 risks. New health conditions were reflected by the question: "Thinking about what has happened since we last saw you, has a doctor ever told you that you developed a new health condition?" The response options were High blood pressure or hypertension, Angina, a heart attack (including myocardial infarction or coronary thrombosis), Congestive heart failure, Diabetes or high blood sugar, A stroke (cerebral vascular disease), Chronic lung disease such as chronic bronchitis or emphysema, Asthma, Arthritis (including osteoarthritis, or rheumatism), Cancer or a malignant tumour (excluding minor skin cancers), Dementia, senility or another serious memory impairment, Alzheimer's disease, and Malignant blood disorder, e.g. leukaemia. Number of new health conditions in wave 1 was distributed as 0 (80.11%), 1 (14.59%), 2 (3.58%), 3 (1.04%), 4 (0.39%), 5 (0.20%), 6 (0.07%), and Here, economic vulnerability included worse financial situations and lowered income. Worse financial situation was reflected by the question: "How do you feel your current financial situation compares to before the coronavirus outbreak?" The response options were "I'm much worse off ", "I'm a little worse off ", "I'm about the same", "I'm a little better off ", and "I'm much better off ". Thus, worse financial situation was recoded as no (=0, "I'm about the same", "I'm a little better off ", and "I'm much better off ") and yes (=1, "I'm much worse off ", "I'm a little worse off "). Lowered income was reflected by the question: "What is your current income level compared to before the coronavirus outbreak before the coronavirus outbreak that began in February?" The response options were higher, about the same, and lower. Here, lowered income was recoded as no (higher/about the same = 0) and yes (lower = 1). Main socio-demographical factors included age (years), gender (male = 1, female = 0), and region (urban = 1, rural = 2) and long-standing condition. For convenience, age was subjectively divided by old group (50-64 years) and older group (65-84 years), oldest group (≥ 85 years). Long-standing condition was reflected by the question: "Do you have any long-standing illness, disability or infirmity?" with the response options of no (= 0) and yes (= 1). First, descriptive analyses were performed by gender in waves 1 and 2. Zero-inflated Poisson regression on COVID-19 risks and new health conditions were performed. Here, COVID-19 risks and new health conditions were count variables. Logistic regressions on healthcare service were examined under the condition that COVID-19 risks and new health conditions were categorized into no (=0) and yes (=1). The study hypotheses from 1a to 2b were statistically similar to models 6 and 29 presented in PRO-CESS macro by reference [69] in supplementary Figs. 1 and 2, respectively. The conditional effect was tested based on a bias-corrected bootstrapping procedure with 5000 samples. A significant effect was signaled by a bootstrap confidence interval (95% CI) that does not include the "0" value. Some cases were deleted due to missing data. The moderating variables were mean centered prior to analysis. Regarding cross-sectional mediation, the mediating effects of COVID-19 risks, worse financial situation, and lowered income on the relationships of "new health conditions → treatment cancellation" and "new health conditions → accessible care" in waves 1 and 2 were conducted by models 6 in Model templates for PROCESS v 2.16 for SPSS (http:// www. guilf ord. com/p/ hayes3). In Model 6, mediators 1, 2, and 3 were COVID-19 risks, worse financial situation, and lowered income, respectively. Cross-sectional moderation models were conducted by model 29 in model templates for PROCESS v 2.16 for SPSS (http:// www. guilf ord. com/p/ hayes3). In Model 29, mediator was COVID-19 risks, W was worse financial situation, V was lowered income. Longitudinal mediation and longitudinal moderation were conducted by models 1 and 2 in MEMORE (Mediation and Moderation for Repeated Measures; available at http:// akmon toya. com), respectively. MEMORE procedure for SPSS version 2.1 was written by Amanda Montoya. COVID-19 risks and new health conditions were continuous variables in cross-sectional mediation, crosssectional moderation, longitudinal mediation, and longitudinal moderation. With respect to data cleaning, the participants for the ELSA COVID-19 Substudy with age < 50 were deleted. In this study, responses with "prefer not to say (-9)", "don't know (-8)", "not asked (-6)", "routing error (-3)", and "item not applicable (-1)" were .0008 .0254 Cnd -> Csk -> Tcn .0001 .0040 .0073 .0010 .0199 Cnd -> Csk -> CFn-> Tcn .0000 .0001 .0001 .0006 Cnd -> Csk -> CIn -> Tcn .0000 .0000 .0001 .0000 .0008 Cnd -> Csk -> CFn-> CIn -> Tcn .0000 .0000 .0000 .0001 .0000 .0004 Cnd -> CFn-> Tcn .0000 Cnd -> Csk-> CIn-> Acr .0001 .0000 .0005 .0000 .0002 Cnd -> Csk -> CFn -> CIn -> Acr .0000 .0000 .0003 .0000 .0001 Cnd -> CFn-> Acr treated as missing values and excluded from the analysis. For this reason, the statistical analysis in this study was performed using original data without imputation. Table 1 showed sample characteristics by gender in waves 1 and 2. There were significant gender differences in age groups, worse financial situation, and lowered income in waves 1 and 2. There were significant gender differences in treatment cancellation and accessible care in wave 1. With regard to income, 16.61% (n = 6930) in wave 1 and 14.09% (n = 6698) in wave 2 were lowered. 19 .89% (n = 6952) suffered new health conditions in waves 1 and 2. In general, there was no statistical significance in gender among all the respondents between regions, COVID-19 risks, and new health conditions in waves 1 and 2. The proportion of treatment cancellation and accessible care was generally high. Some patients with specific health conditions could be at risk of being inappropriately treated due to COVID-19 pandemic. In Table 2 , older adults in age group 65-84 years were less likely to be involved in COVID-19 risks and new health conditions than adults in age group 50-64 years in waves 1 and 2. Older adults in age group ≥85 years were less likely to be involved in COVID-19 risks (wave 1) and new health conditions (waves 1 and 2) than adults in age group 50-64 years. Males had a lower probability of being involved in COVID-19 risks and new health conditions than females. Older adults with long-standing condition were more likely to be involved in new health conditions (in waves 1 and 2) and more likely to be involved in COVID-19 risks (in wave 2) than without. Older adults with lowered income were less likely to suffer new health conditions than those without in waves 1 and 2. In Table 3 , older adults in age group 65-84 years had lower probability of treatment cancellation in waves 1 and 2 and accessible care in wave 1 and higher probability of accessible care in wave 2 as compared with older adults in age group 50-64 years. Compared with older adults in age group 50-64 years, older adults in age group ≥85 years had lower probability of treatment cancellation in waves 1 and 2 and higher probability of accessible care in wave 2. Compared with females, males had lower probability of treatment cancellation in waves 1 and 2 and accessible care in wave 1 and higher probability of accessible care in wave 2. Compared with older adults in urban place, older adults in rural place had lower probability of treatment cancellation in waves 1 and 2 and higher probability of accessible care in wave 2. Older adults with long-standing condition were more likely to cancel treatment and access care than without (in wave 2). Compared with older adults without coviod-19 risk, Table 6 Longitudinal mediations Tcn1 treatment cancellation in wave 1, Tcn2 treatment cancellation in wave 2, Acr1 accessible care in wave 1, Acr2 accessible care in wave 2, Csk1 COVID-19 risks in wave 1, Csk2 COVID-19 risks in wave 2, CFn1 worse financial situation in wave 1, CFn2 worse financial situation in wave 2, CIn1 lowered income in wave 1, CIn2 lowered income in wave 2 older adults with coviod-19 risk had lower probability of treatment cancellation in waves 1 and 2 and higher probability of accessible care in wave 2. Older adults with condition were more likely to cancel treatment and access care than without, while older adults with financial were less likely to cancel treatment and access care than without (in wave 1). Older adults with lowered income were less likely to cancel treatment (in wave 1) and more likely to access care than without (in wave 2). In Table 4 In Table 5 , the coefficients of Cnd × CFn, Csk × CIn, Cnd × CFn, and Cnd × CIn were not significant in wave 1. Thus, there were no moderating effects of worse financial situation and lowered income on the mediating COVID-19 risks on the associations of new health conditions with treatment cancellation. Likewise, there were no moderating effects of worse financial situation and lowered income on the mediating COVID-19 risks on the associations of new health conditions with accessible care. Furthermore, the coefficients of Cnd × CFn were significant in wave 2. But, the coefficients of Csk × CIn, Cnd × CFn, Cnd × CIn were not significant. Thus, there were moderating effects of worse financial situation on the Cnd → Csk. Accordingly, H2a and H2b were rejected. As presented in Table 6 Consequently, H3a and H3b were rejected. As presented in Table 7 The proportion of treatment cancellation and accessible care in wave 2 was higher than those in wave 1. Empirically, zero-inflated Poisson regression and logistic regression indicated COVID-19 risks, worse financial situation, and lowered income were associated with treatment cancellation and accessible care. Obviously, older adults with worse financial situation and lowered income had a considerably less likelihood of treatment cancellation and accessible care in wave 1 rather than 2. The current research found the effects of new health conditions on treatment cancellation and accessible care indirectly through worse financial situation, lowered income and COVID-19 risks existed in longitudinal mediation models rather than cross-sectional mediation models, cross-sectional moderation models, and longitudinal moderation models. Thus, COVID-19 risks, worse financial situation, and lowered income were most identified as barriers to healthcare in this study. The results in this study were in line with the findings in a study among adult Israeli Jews. The study reported adults aged ≥60+ years and reporting higher levels of COVID-19 fear were more likely to report forgone care than younger and less concerned adults during the COVID-19 lockdown period [70] . Likewise, COVID-19-associated delays in seeking care were reported in Antananarivo, Madagascar [71] . Similarly, a cross-sectional study indicated that the COVID-19 pandemic was associated with reductions in hospitalizations for ambulatory care-sensitive conditions [72] . The findings from longitudinal mediation could be partially explained by some studies. Serious adverse consequences accompanied by COVID-19 restrictions are reported in the field of health management. Inaccessible health care among adults with diabetes during the COVID-19 pandemic was documented in the United States [73] . COVID-19 lockdowns lead to declines in health care utilization. It is necessary to equip community health care services amid COVID-19 [74] . Likewise, insignificant cross-sectional mediation models, cross-sectional moderation models, and longitudinal moderation could also be explained by findings from international comparison study. The study indicated negative financial effects on health care providers in the England were not severe than that in the United States during COVID-19 pandemic [75] . The findings in this study were in line with the mismatch between health care demand and supply reported in early studies. For example, there were high disproportionate risks for COVID-19 hospitalization pressures varied by spatial and socio-demographic factors in England and Wales [76] . A qualitative study found that fear of COVID-19 was a major factor to hinder access to maternal healthcare services during the onset of the COVID-19 pandemic in Kenya [77] . A Chinese study indicated social factors and type of healthcare utilization were associated with the time delay for diagnosis [78] . Furthermore, an observational cohort study indicated worse health and function were partial determinants of more hospital admissions among people aged ≥50 years with lower wealth [79] . Also, a repeated cross-sectional study indicated older adults experiencing financial hardship had markedly greater risks of mental disorders after accounting for COVID-19 social assistance receipt [80] . This research is helpful in gaining a better understanding of the relationship between health service needs and health care provision. The role of economic conditions was explanatory in line with findings in prior studies. In details, unmet need in health care use may relate to individuals' ability to pay [81] . Additionally, poorer access to care might be the determinants of higher levels of unmet social needs among low-income adults [82] . Especially, low-income individuals with serious health conditions have poorer access to outpatient care than their higher income counterparts [83] . Some conjecture may be correct. COVID-19 risks, worse financial situation and lowered income must exacerbate adults with long-standing condition and new health conditions face accessing health care. Some people with long-standing condition and new health conditions are at great risk of contracting COVID-19 because they require health and social care and are unable to access. Additionally, some people with worse financial situation and lowered income have COVID-19 risks if they become infected. Despite this, UK governments have recognized and responded to diverse health care needs of people with long-standing condition, new health conditions, worse financial situation and lowered income during COVID-19 pandemic. As for policy practice, the findings of the current study indicated older adults at risk for experiencing worse financial situation and lowered income could be targeted to participate in interventions to reduce their levels of COVID-19 risks. The fundamental role of COVID-19 risks as a mediator makes it evident that reduction in COVID-19 risks can promote health-seeking behaviors. The findings stress the importance of developing interventions aimed at mitigating COVID-19 risks, worse financial situation and lowered income while consuming healthcare services. In the short term, healthcare services need to reduce COVID-19 risks under conditions of uncertain demand. In the long term, policymakers should consider alternative ways of delivering healthcare services to the adults with worse financial situation and lowered income during crisis. Therefore, there is a need to redesign the strategies of health care service during COVID-19 pandemic. Additionally, it is necessary to strengthen primary health care services, particularly by diminishing obstacles for groups with worse financial situation and lowered income. Persons with preexisting health conditions or disabilities may be worse due to the COVID-19 pandemic controlling strategies. A study reported that disruption to care caused by COVID-19 or insufficient quality of the new telemedicine care could lead to psychological disorders which was mediated by self-efficacy [84] . Early screening, recognition, diagnoses, and treatment of health conditions can prevent complications, improve quality of life, and help reduce health care costs. Untreated medical conditions affect mortality, functional disability, quality of life, and health care costs, especially for the uninsured individuals. Part of patients with specific health conditions could be at risk of being inappropriately treated during the COVID-19 pandemic. Family members can play in supporting patients' selfcare. MHealth can reduce barriers to health care access and facilitate integrated health care models for vulnerable populations [85] . With respect to indirect impact of the COVID-19 pandemic on the health-care system, a retrospective cohort study indicated undiagnosed conditions or delayed diagnosis of health conditions increased substantially [86] . Some limitations of the present study should be noted. First, a short-term longitudinal study with two waves could not draw accurate causal conclusions. Second, some key variables like lockdown, lifestyle and treatment cost were not surveyed in the dataset used in this study. Similarly, some key variables like medical counseling in wave 1 did not appear in wave 2. Particularly, limited to the topic of interest, some vital independent variables were not inserted into regression models. For example, ageing can make adults frail and depend on transport to health access of the elderly. But, there were no indirect association between age and health care use to report in this study. Finally, low response rates in possibly limited the generalization of the empirical findings. Thus, the findings in this study cannot be generalised to the wider communities in these countries. First, mediation and moderation analyses were performed after regression screen out potential mediators and moderators. Second, mediation and moderation analyses of COVID-19 risks, new health conditions, and worse financial situations on the relationships of new health conditions → treatment cancellation and new health conditions → accessible care were conducted using the PROCESS tool and MEMORE procedure. First, association of functional status with treatment cancellation and accessible care need to be explored in the future. Second, association of mental status with treatment cancellation and accessible care may be new direction in the future. This is can be reflected by several studies which reported the association between COVID-19-related financial situation and mental health [87, 88] . If replicating in future studies, it is recommended that research with a longitudinal design be undertaken irrespective of crisis settings. Moreover, more research is needed to explore the associations of interest with other mediators and moderators like transportation barriers. In summary, the findings in longitudinal mediation models reveal the potential influence of worse financial situation, lowered income and COVID-19 risks and their relevance for future research. There were not significant mediating and moderating worse financial situation, lowered income and COVID-19 risks in the relationships: new health conditions → treatment cancellation and new health conditions → accessible care. The COVID-19 risk negatively mediated treatment cancellation and accessible care, while worse financial situation and lowered income positively mediated them. Abbreviations COVID-19: Coronavirus disease-19; IRR: Incidence relative risk; 95% CI: 95% confidence interval; Csk: COVID-19 risks; CFn: Worse financial situation; CIn: Lowered income; Cnd: New health conditions; Tcn: Treatment cancellation; Acr: Accessible care; LLCI: Low limit confidence interval; ULCI: Upper limit confidence interval. The online version contains supplementary material available at https:// doi. org/ 10. 1186/ s12877-022-02995-8. Erectile Dysfunction and Treatment: An Analysis of Associated Chronic Health Conditions Traumatic Brain Injury and Treatment of Behavioral Health Conditions Treatment challenges and health conditions among people living with HIV with or without substance use disorder in the Russian Federation Comorbid Health Conditions and Treatment Utilization among Individuals with Opioid Use Disorder Experiencing Homelessness Oral Health Conditions and Treatment Needs of Children with Intellectual Disabilities Attending Special Olympics Australia Effectiveness of treatment for 6813 patients with mental health conditions in Cambridgeshire: a cross-sectional study Perceived Treatment Satisfaction and Effectiveness Facilitators Among Patients With Chronic Health Conditions: A Self-Reported Survey Barriers and facilitators for treatment-seeking for mental health conditions and substance misuse: multi-perspective focus group study within the military Self-reports of treatment for secondary health conditions: results from a longitudinal community survey in spinal cord injury Effects of the Affordable Care Act on Private Insurance Coverage and Treatment of Behavioral Health Conditions in Young Adults Mental Health Conditions Among Community College Students: A National Study of Prevalence and Use of Treatment Services Collaborative care in the treatment of opioid use disorder and mental health conditions in primary care: A clinical study protocol Integrating Treatment for Co-Occurring Mental Health Conditions Problem gambling, associations with comorbid health conditions, substance use, and behavioural addictions: Opportunities for pathways to treatment Health Seeking Behavior Among Adults and Elderly With Chronic Health Condition(s) in Albania. Front Public Health Hypertension Status and Associations with Self-Rated Health and General Practitioner Health Seeking in a Rural Australian Cohort Diagnosis and Care of Chronic Health Conditions Among Medicaid Expansion Enrollees: a Mixed-Methods Observational Study Stigma is leading to under-treatment of mental health conditions, says leading psychiatrist Treatment costs in Canada of health conditions resulting from chronic hepatitis B infection Narrating uncertainties about treatment of mental health conditions Living conditions, ability to seek medical treatment, and awareness of health conditions and healthcare options among homeless persons in Disparities in the reporting and treatment of health conditions in children: an analysis of the Medical Expenditure Panel Survey The health conditions and the health care consumption of the uninsured Dynamic treatment regimes for managing chronic health conditions: a statistical perspective Personalised care planning for adults with chronic or long-term health conditions Assessment of self-care and medication adherence in individuals with mental health conditions Barriers, facilitators, and proposals for improvement in the implementation of a collaborative care program for depression: a qualitative study of primary care physicians and nurses Acceptance and commitment-based therapy for patients with psychiatric and physical health conditions in routine general hospital care -Development, implementation and outcomes Health Equity in the Effectiveness of Web-Based Health Interventions for the Self-Care of People With Chronic Health Conditions: Systematic Review Early impacts of the COVID-19 pandemic on mental health care and on people with mental health conditions: framework synthesis of international experiences and responses Exploring the Complex Pathways between the Fear of COVID-19 and Preventive Health Behavior among Nigerians: Mediation and Moderation Analyses COVID-19 and Social Isolation Endangering Psychological Health of Older Adults: Implications for Telepsychiatry Testing the Multi-Theory Model (MTM) to Predict the Use of New Technology for Social Connectedness in the COVID-19 Pandemic Accessibility and utilisation of telehealth services among older adults during COVID-19 pandemic in the United States. Health Soc Care Community Effects of the COVID-19 pandemic on the follow-up and treatment of patients with idiopathic pulmonary fibrosis: a cross-sectional, multicentre phone call survey Impact of COVID-19 on gastric cancer treatment in Japanese high-volume centers: a JCOG stomach cancer study group survey Disease monitoring of biologic treatment in IBD: early impact and future implications of COVID-19 pandemic The impact of the COVID-19 pandemic on asthma treatment in Japan: Perspectives based on doctors' views Orthodontic treatment pause during COVID-19 outbreak: Are we overlooking potential harms to our patients and their treatment outcomes? Management of bone metastasis and cancer treatment-induced bone loss during the COVID-19 pandemic: An international perspective and recommendations Patients with Chronic Pain Prefer Maintenance of Pain Treatment Despite COVID-19 Pandemic Restrictions Life on pause: An analysis of UK fertility patients' coping mechanisms after the cancellation of fertility treatment due to COVID-19 Emotional Impact of Delay in Fertility Treatment due to COVID-19 Pandemic The psychological impact of fertility treatment suspensions during the COVID-19 pandemic The impact of the COVID-19 pandemic on the emotional well-being and home treatment of Belgian patients with cystic fibrosis, including transplanted patients and paediatric patients Applying the Health Belief Model to Characterize Racial/Ethnic Differences in Digital Conversations Related to Depression Pre-and Mid-COVID-19 Determinants of acceptance of Coronavirus disease-2019 (COVID-19) vaccine among Lebanese health care workers using health belief model Acceptance of seasonal influenza vaccination and associated factors among pregnant women in the context of COVID-19 pandemic in China: a multi-center cross-sectional study based on health belief model Negative impacts of COVID-19 lockdown on mental health service access and follow-up adherence for immigrants and individuals in socio-economic difficulties Assessing Income-Related Inequality on Health Service Utilization among Chinese Rural Migrant Workers with New Co-Operative Medical Scheme: A Multilevel Approach Association of household income with home-based rehabilitation and home help service utilization among long-term home care service users Financial Barriers to Access to Health Services for Adult People with Disability in Iran: The Challenges for Universal Health Coverage Removing financial barriers to access reproductive, maternal and newborn health services: the challenges and policy implications for human resources for health The impact of reducing financial barriers on utilisation of a primary health care facility in Rwanda Community based saving groups: an innovative approach to overcome the financial and social barriers in health care seeking by the women in the rural remote communities of Pakistan. Arch Public Health Forgone care and financial burden due to out-of-pocket payments within the German health care system Social inequalities in self-reported refraining from health care due to financial reasons in Sweden: health care on equal terms? BMC Financial hardship and health risk behavior during COVID-19 in a large US national sample of women Financial hardship, unmet medical need, and health self-efficacy among African American men Health seeking for chronic lung disease in central Malawi: Adapting existing models using insights from a qualitative study Health-seeking behavior and barriers to treatment of patients with upper gastrointestinal cancer detected by screening in rural China: real-world evidence from the ESECC trial Understanding health-seeking and adherence to treatment by patients with esophageal cancer at the Uganda cancer Institute: a qualitative study Health-Seeking Behaviour and Delayed Presentation of Oral Cancer Patients in a Developing Country: A Qualitative Study based on the Self-Regulatory Model Quality of Life and Illness Acceptance among End-Stage Renal Disease (ESRD) Patients on Hemodialysis: The Moderating Effect of Death Anxiety during COVID-19 pandemic Relationship between selfcare activities, stress and well-being during COVID-19 lockdown: a crosscultural mediation model Dealing With COVID-19 Patients: A Moderated Mediation Model of Exposure to Patients' Death and Mental Health of Italian Health Care Workers Self-efficacy and fatigue among non-frontline health care workers during COVID-19 outbreak: A moderated mediation model of posttraumatic stress disorder symptoms and negative coping A chain mediation model on COVID-19 symptoms and mental health outcomes in Americans Introduction to mediation, moderation, and conditional process analysis Prevalence and correlates of forgone care among adult Israeli Jews: A survey conducted during the COVID-19 outbreak Surge of Typhoid Intestinal Perforations as Possible Result of COVID-19-Associated Delays in Seeking Care Trends in Hospitalizations for Ambulatory Care-Sensitive Conditions During the COVID-19 • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations maximum visibility for your research: over 100M website views per year submit your research ? Choose BMC Health Care Access and Use Among Adults with Diabetes During the COVID-19 Pandemic -United States Associations between accessibility to health care service, social support, and Korean Americans' mental health status amid the COVID-19 pandemic The Structure and Financing of Health Care Systems Affected How Providers Coped With COVID-19 Forecasting spatial, socioeconomic and demographic variation in COVID-19 health care demand in England and Wales Health-seeking behaviour among pregnant women during the COVID-19 pandemic: A qualitative study Health-seeking and diagnosis delay and its associated factors: a case study on COVID-19 infections in Shaanxi Province Socioeconomic position and use of hospital-based care towards the end of life: a mediation analysis using the English Longitudinal Study of Ageing Financial hardship and social assistance as determinants of mental health and food and housing insecurity during the COVID-19 pandemic in the United States Unmet health care need and income-Related horizontal equity in use of health care during the COVID-19 pandemic Unmet social needs among low-income adults in the United States: Associations with health care access and quality Income Related Inequality of Health Care Access in Japan: A Retrospective Cohort Study Disruptions in the management and care of university students with preexisting mental health conditions during the COVID-19 pandemic Diagnosis of physical and mental health conditions in primary care during the COVID-19 pandemic: a retrospective cohort study Integrated Health Care and mHealth: A Model of Care for Refugees with Complex Health Conditions COVID-19-Related Financial Hardship Financial distress and suicidal behaviour during COVID-19: Family identification attenuates the negative relationship between COVID-related financial distress and mental Ill-health Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations The authors of this paper would like to acknowledge the very helpful comments of three reviewers on the original submission. BXH designed the study, performed the statistical analysis, and completed the original version. HYG performed Tables 1, 2 , 3, 4, 5, 6 and 7 under the tutorship of BXH. The authors read and approved the final manuscript. Bingxue Han is an associate professor of College of Urban and Environmental Sciences, Xuchang University. Han also is a member of the Family Issues Center at Xuchang University. Her academic interest is rural development, rural education, and rural health in modern China. Access to the survey data is open and publicly available in the following link. https:// www. elsa-proje ct. ac. uk/ covid-19 The data adopted was from a publicly available survey dataset whose ethical approval was obtained from the institutional review board at The English Longitudinal Study of Ageing. All methods were carried out in accordance with relevant guidelines and regulations. Written informed consent was obtained from all participants before they agreed to participate in the study. Participants were informed that they could leave the study at any time without penalty, and all personal information was kept confidential. Thus, it was not necessary to obtain ethical approval from the institutional review board at the author's institution. Not applicable. The authors declare that they have no competing interests regarding the publication of this paper.