key: cord-0821068-1uh1sl9n
authors: Newbronner, E.; Spanakis, P.; Wadman, R.; Crosland, S.; Heron, P.; Johnston, G.; Walker, L.; Gilbody, S.; Peckham, E.
title: Business as Un-usual: Access to mental health and primary care services for people with severe mental illness during the COVID-19 restrictions
date: 2021-05-25
journal: nan
DOI: 10.1101/2021.05.24.21257694
sha: 39e158c1ab1d3c332a4f702e62b2ad085b7a9b1b
doc_id: 821068
cord_uid: 1uh1sl9n

Aims: To explore: how satisfied people with severe mental illness (SMI) are with the support received during the pandemic; understand any difficulties encountered when accessing both mental health and primary care services; consider ways to mitigate these difficulties; and assess the perceived need for future support from mental health services. Materials and Methods: A representative sample was drawn from a large transdiagnostic clinical cohort of people with SMI, which was recruited between April 2016 and March 2020. The sample was re-surveyed a few months after the beginning of the restrictions. Descriptive frequency statistics were used to analyse the quantitative data. The free text responses were analysed thematically. Results: 367 participants responded to the survey. Two thirds were receiving support from mental health services with the rest supported in primary care or self-managing. A quarter thought they would need more mental health support in the coming year. Half had needed to used community mental health services during the pandemic and the majority had been able to get support. A minority reported that their mental health had deteriorated but they had either not got the supported they wanted or had not sought help. The biggest service change was the reduction in face-to-face appointments and increasing use of phone and video call support. Nearly half of those using mental health services found this change acceptable or even preferred it; acceptability was influenced by several factors. Participants were more likely to be satisfied with support received when seen in person. Discussion: Although most participants were satisfied with the mental health support they had received, a minority were not. This, couple with findings on future need for mental health support has implications for post pandemic demand on services. Remote care has brought benefits but also risks that it could increase inequalities in access to services.

In mid-March 2020, soon after the WHO declared the outbreak of COVID-19 to be a 49 pandemic, sweeping changes in the way mental health and primary care services in the UK 50 are provided were introduced. A survey of mental health staff, conducted early in the first 51 wave of the pandemic in the UK (i.e April/May 2020) found that the number of face-to-face 52 meetings was reduced, and phone and video appointments and support were rapidly 53 introduced [1] . This observation was supported by a recent report from the House of Lords 54

Covid-19 Committee [2] . In most areas of the UK the frequency of contacts and the range of 55 both NHS and third sector services available also reduced [3] . Research is beginning to 56 explore the impact of the pandemic and the pandemic restrictions on people with pre-57 existing mental health conditions. However, there has been little research into the specific 58 experiences of people living with severe mental illness (SMI), and in particular how their 59 access to mental health and primary care services may have changed. 60 61 People with SMI are an especially vulnerable group who already experience significant 62 health inequalities. Notably they currently experience a mortality gap of 15-20 years when 63 compared to people without SMI [4, 5, 6 ]. An important driver for this gap is preventable 64 physical health conditions, linked to behavioural risk factors, such as poor diet and smoking. 65

However, access to and take-up of services is also a factor [7, 8] . There are concerns that 66 the changes in the way services have been delivered during the pandemic may have further 67 increased the barriers to access that people with SMI experience (e.g. access to digital 68 technologies or familiarity/confidence in using them) [9,10]. It is therefore important to: 69 explore how satisfied people with SMI are with the support they received during the 70 pandemic; understand any difficulties they encountered when accessing both mental health 71 and primary care services; consider ways to mitigate these difficulties; and assess the 72 perceived need for future support from mental health services. The aim of this study was to 73 use a large clinical cohort of people with SMI, which was recruited in the years immediately 74 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) to rapidly identify participants during the COVID-19 pandemic. To increase the chances of 97 participants responding and their contact details remaining current, we invited the 98 participants who had been more recently recruited to the CtG cohort. People who met the 99 eligibility criteria were contacted by telephone or letter and invited to take part in the OWLS 100 COVID-19 study. Those who agreed to take part were offered three options; i) complete the 101 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 25, 2021. ; https://doi.org/10.1101/2021.05.24.21257694 doi: medRxiv preprint 5 survey over the phone with a researcher, ii) be sent a link to complete the survey online or 102 iii) be sent a hard copy of the questionnaire in the post to complete and return. 103

In Section 2 of the survey, we explored how participants' access to mental health and 105 primary care services had changed during the pandemic. For example, we asked 106 participants who were receiving support from mental health services, whether they had 107 experienced one or more specific change (e.g. seeing a different mental health worker to the 108 person they normally saw). If they had, they could indicate how they felt about the change 109 with one of three answers (i.e. I like it better; its ok -not better or worse; or I don't like it). 110

The full OWLS 1 questionnaire is shown in the supplementary materials (S1). For all 111 services, we were interested to understand how changes in the way services were provided, 112 in particular the shift to phone and online meetings and appointments, might have affected 113 peoples' experiences. We also wished to explore what factors might have hindered or 114 facilitated good care and support, despite the change in access to services. For mental 115 health services we also asked people about the support they anticipated needing from 116 services in the year ahead, an issue of particular interest to NHS providers. 117

The study analysis plan was registered on Open Science Framework (available at 119 https://doi.org/10.17605/OSF.IO/E3KDM -section 2.1). Descriptive frequency statistics (Ns 120 and percentages) were used to describe key sample characteristics and service use 121 variables. 2x2 cross-tabulations and chi-square tests were used to examine the association 122 between a) unmet need for support and changes in mental or physical health, and b) mode 123 of service delivery and service satisfaction. The latter was examined for each service type 124 separately (GPs, Community mental health, Mental health crisis services) and p values were 125 corrected for multiple testing (p multiplied by four). For brevity, the results for general 126 hospital services are not included in the main text of the paper but are available in the 127 supplementary materials (S2). The statistical significance criterion used for all analyses was 128 p < 0.05. Analyses were undertaken using SPSS v.26. 129 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In the pre-registered analysis plan it was stated that mode of service delivery would be 130 coded at three-levels (in person, over the phone, or online e.g. video call). However, too few 131 participants reported receiving services online (N ranged from 2 to 14, depending on service 132 type) to allow for meaningful comparisons, and so this was merged with receiving services 133 over the phone into a remote delivery category. The plan also stated that association 134 between Community mental health service satisfaction and mode of service delivery would 135 be stratified per primary or secondary care patients. As very few participants currently in 136 primary care reported getting support from Community mental health services (N = 24) to 137 allow for meaningful comparisons, analysis was conducted only in the full sample. 

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 25, 2021. ; https://doi.org/10.1101/2021.05.24.21257694 doi: medRxiv preprint 7 In this section we begin by briefly describing the demographic characteristics of the OWLS 156 participants. We then present our findings in relation to participants experiences of 157 accessing both community mental health services and mental health crisis services, and 158 their perceived need for future support from mental health services. We then go on to 159 describe their experiences of using primary care services. 160 161

The first survey for the OWLS COVID-19 study, which was conducted between July to 162 December 2020, recruited 367 people from the CtG Cohort. Table 1 describes the socio-163 demographic characteristics and diagnosis of OWLS participants. It should be noted that it 164 was an ethical requirement of the CtG study that the participants consent to their diagnosis 165 being provided to the research team and some participants did not consent to this. In the 166 interests of inclusivity and because there may be some differences between those who 167 consented to their diagnosis being provided and those who did not, we did not make it a 168 requirement of OWLS that the participants should consent to their diagnosis being provided. 169

The mean age was 50.5 (range = 20 to 86, SD ± 15.69) with 51.0% male and 77.4% white 170

British. 171 We examined whether there was any difference between those who said that their mental 190 health had deteriorated and those who said it had not, in terms of getting support. As Table 2  191 below shows there was no significant association between deterioration in mental health and 192 receiving mental health services (χ 2 (1) = .99, p = .319). However, 14.5% (n=12) did report 193 that their mental health had deteriorated but they had either not got the supported they 194 wanted or had not sought help. One of the free text comments illustrates this: "Before the 195 pandemic I wasn't using MH services but as my MH has gotten worse then I have become 196 very conscious of not being able to access services". 197 The pandemic led to substantial and rapid changes in the way in which mental health 208 services were provided. The biggest shift was the reduction in face-to-face appointments 209 and increasing use of phone and online consultations and support. However, people also 210 experienced less frequent contact with services, or had a more limited range of services 211 (including community and voluntary sector services) available to them. A few people saw a 212 different mental health worker. We asked the people who were currently receiving support 213 from mental health services (n=224) how they felt about these changes and 221 provided 214 responses. Table 3 provides and overview of their responses. 215 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 25, 2021. The results show that despite the impact of the pandemic, a high proportion of those survey 219 respondents who were in contact with mental health services, were seeing the same mental 220 health worker, often in the same place. Clearly many people were being supported on the 221 telephone or online instead of face to face, and whilst almost a third did not like this change, 222 nearly half found it acceptable or even preferred it. 223

The findings from the OWLS study in relation to access to digital devices are reported in a 225 separate paper. However, it is useful to note here that the majority of study participants did 226 have access to a smartphone, tablet, laptop and/or desktop computer. However, 13.4% 227 (n=49) did not have any of these devices. The free text comments also suggested that even 228 where people did have access to such devices, they may not have exclusive use (e.g. a 229 'family' computer), their internet access may be limited or there may be practical and 230 emotional concerns associated with online appointments. 231 232 Focusing specifically on the 168 people who had received support from community mental 233 health services, just over half (n=89/52.7%) had received that support in person. Sixty-six 234 people (39.1%) had support over the phone and a minority (n=14/3.8) had some kind of 235 online video support (e.g. NHS Attend Anywhere). Almost two thirds (n=112/65.9%) were 236 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 25, 2021. ; https://doi.org/10.1101/2021.05.24.21257694 doi: medRxiv preprint completely satisfied with the support they received but as Table 4 below shows, there was a 237 significant association between satisfaction with the support received from community 238 mental health services and the way in which it was provided (χ 2 (1) = 22.92, p < .001). 239

Specifically, the proportion of those who reported receiving support in person and were 240 completely satisfied was significantly higher that the proportion of those who reported 241 receiving support remotely and were completely satisfied. 242 face visits and built a rapport so that now that we speak online it is okay -without the face-257 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Over three quarters of this group (n=45/76.3%) were able to get support but a significant 275 minority (n=14/23.7%) reported that they either did not get the support they needed or did 276 not try to get help. The free text comments suggest that for this group the implications of not 277 being able to access support were potentially very serious: We also explored satisfaction with the way in which support was provided (see Table 5 ). We 286 found there was no significant association between receiving the level of support needed 287 from mental health crisis services and the way in which this support was provided (χ 2 (1) = 288

.07, p = 4.000) (i.e. the proportion of those who were completely satisfied with the support 289 they received was broadly similar for the in person and phone/online groups). 290 

There is substantial interest within the NHS about the impact of the COVID-19 pandemic on 295 future need for mental health services and we were able to explore this to a limited extent in 296 this study. We asked those people who were not currently getting support from mental health 297 services whether they thought they would need support in the year ahead, and just under a 298 third (n=44/31.7%) thought they would. 299 300 For those people who were already being supported by mental health services, we asked 301 how their need for support might change in the year ahead. As Table 6 below shows, just 302 over a quarter thought they would need more support either because their mental health had 303 declined, or they had been putting off dealing with some issues, or they felt the support they 304 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 25, 2021. ; https://doi.org/10.1101/2021.05.24.21257694 doi: medRxiv preprint 15 had prior to the pandemic was insufficient, and almost a quarter thought they might need 305 more support. 306 Table 6 Perceived need for support from mental health services in the year ahead 307

About the year ahead, which of the following fits you best? during the pandemic (for mental and/or physical health conditions). Of this group, the 313 majority (n=207/86%) reported that they had been able to get an appointment and most 314 (n=154/74.4%) felt they had received the care and support they needed. However, like 315 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 25, 2021. ; https://doi.org/10.1101/2021.05.24.21257694 doi: medRxiv preprint mental health services, GP practices also moved to much greater use of phone and video 316 consultations. We examined whether there was any association between deterioration in 317 physical health and access to GP care and support. Table 7 below shows that there was no  318 significant association (χ 2 (1) = .14, p = .706). 319 We then examined whether people felt they had received the support they needed. Almost 323 three quarters reported that they had but there was a marked different between those who 324 had had a face-to-face appointment and those supported on the phone or online. As Table 8  325 shows, people were more likely to feel completely satisfied with the support they received 326 when it was face-to-face (χ 2 (1) = 10.46, p = .008). 327 In relation to medication and pharmacy services, most survey respondents (n=330/89%) had 331 needed prescription medication during the pandemic and almost all were able to obtain their 332 medication from their usual pharmacy. However, the free text comments revealed that a few 333 people had experienced delays and difficulties with the supply of their usual medication. 334

One respondent said that they were on three types of medication, and at one point had to go 335 to three different pharmacies to get the medication they needed. They noted that they were 336 relatively well but felt that if they had not been, they might have stopped taking their 337 medication. Another explained: 338 339 "There was a shortage of medication supply in the area. I struggled to get my 340 prescription tablets and when I eventually got them then I got a slightly different type 341

with them -at the time I thought they were going to give me a different drug and the 343 ones I take are quite a fine balance. So that was very worrying." 344 345

Most participants reported that they had been able to access support from community 347 mental health services and a relatively high proportion said they were satisfied with the 348 support they received. This contrasts with the findings from a qualitative study by Gillard et al 349 [10] in which participants reported issues around continuity of care, not getting treatment as 350 usual and service changes. However, in our study a minority did report that their mental 351 health had deteriorated but they had either not got the supported they wanted or had not 352 sought help. This is an issue of concern, which alongside our findings on future need for 353 increased support, may have implications in terms of post-pandemic increased demand for 354 mental health services. Furthermore, for some people living with severe mental illness, the 355 fear of becoming unwell and not being able to access support is a source of anxiety. A 356 substantial minority of participants in our study did not feel confident that support would be 357 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted May 25, 2021. ; https://doi.org/10.1101/2021.05.24.21257694 doi: medRxiv preprint

Impact on Mental Health Care and on Mental Health Service Users of the COVID-19

Pandemic: A Mixed Methods Survey of UK Mental Health Care Staff

Beyond Digital: Planning for a Hybrid World. 1st Report of the 427 session 2019-21 House of Lords COVID-19 Committee

 417 We would like to thank all the participants in the OWLS study, the lived experience panel 418 who provided advice throughout the study, and the mental health trust and CRN staff who 419 supported study recruitment. 420 421

available should they need -a perception that mental health services may need to explore 358 and address. 359 360 The biggest change in the way in which mental health services were provided was the 361 reduction in face-to-face appointments and the increase in remote care. We found that 362 almost half those who had used community mental health service had received support in 363 person, with staff using creative approaches to maintaining face-to-face contact. Many 364 people were being supported on the telephone and a few had online support. Whilst almost 365 a third did not like this change, nearly half found it acceptable or even preferred it. However, 366the free text comments suggest that acceptability was influenced by several factors. In 367 particular, respondents felt that telephone or video appointments were far easier when: they 368 had an established relationship with their mental health professional; where they had 369 somewhere private to hold a telephone conversation; or they had access to a (private) digital 370 device -findings which echo those of a recent qualitative study by Liberati et al [13] . 371 Furthermore, whilst increased use of remote mental health care might be acceptable for a 372 limited period, during a public health crisis, in the long term it could lead to problems with 373 accessibility, equity of care and service quality [9,10,13]. The limited availability of online 374 support reported by participants in our study, suggests that Greenhalgh et al [14] are correct 375 in suggesting that the evidence relating to video-based remote care needs to be 376 strengthened. Interestingly, whilst people were more likely to be satisfied with the support 377 they received from community mental health service when they had been seen in person, in 378 mental health crisis services the way support was provided appeared less important. We 379 speculate that this could be because phone support is already commonly used in crisis 380 services and/or because remote care facilitated more rapid access to support. 381 382 With regard to access to primary care services, the majority of participants had been able to 383 get an appointment and most felt they had received the care and support they needed. 384 However, those who had a face-to-face appointment were more likely to feel satisfied with 385 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted May 25, 2021. ; https://doi.org/10.1101/2021.05.24.21257694 doi: medRxiv preprint 19 the support they received. Interestingly, the free text comments revealed a specific issue in 386 relation to obtaining medication, with a number of respondents reporting delays and 387 difficulties with the supply of their usual medication. 388 389 Overall then, our study found that the majority of people with SMI were able to access 390 support but those who received remote support were less likely to be satisfied with that 391 support. Furthermore, the context of remote support was important. This suggests that if, as 392 seems likely, service providers continue to employ some element of remote care, service The findings from this study suggest that some people with SMI do not feel that they have 401 received enough support during the pandemic, or have been reluctant to seek support, 402 perhaps because of the way services were provided or fear of contracting the virus. As a 403 consequence, when the pandemic restrictions ease there could be an increase in demand 404 from mental health services from existing service users. However, this may be tempered by 405 the fact that many mental health professionals and GPs appear to have prioritised support 406 for people with SMI and found creative ways to maintain some face-to-face contact. 407 408 Some of the changes introduced as a result of the COVID-19 pandemic will almost certainly 409 continue to be part of future mental health and primary care service. These changes can 410 bring benefits for service providers and some service users. However, they could also 411 increase the risk of unequal access to support and poorer quality of care. To counter these 412 risks, people with SMI must be consulted about the nature of any long-term changes, and 413 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)The copyright holder for this preprint this version posted May 25, 2021. ; https://doi.org/10.1101/2021.05.24.21257694 doi: medRxiv preprint