Article Text
Abstract
Introduction Infection prevention and control (IPC) is imperative towards patient safety and health. The Infection Prevention and Control Assessment Framework (IPCAF) developed by WHO provides a baseline assessment at the acute healthcare facility level. This study aimed to assess the existing IPC level of selected public sector hospital facilities in Punjab to explore their strengths and deficits.
Methods Between October and April 2023, 11 public sector hospitals (including tertiary, secondary and primary level care) were selected. Data were collected using the IPCAF assessment tool comprising eight sections, which were then categorised into four distinct IPC levels– inadequate, basic, intermediate and advanced. Key performance metrics were summarised within and between hospitals.
Results The overall median IPCAF score for the public sector hospitals was 532.5 (IQR: 292.5–690) out of 800. Four hospitals each scored ‘advanced’ as well as ‘basic’ IPC level and three hospitals fell into ‘intermediate level’. Most hospitals had IPC guidelines as well as IPC programme, environments, materials and equipments. Although 90% of secondary care hospitals had IPC education and training, only 2 out of 5 (40%) tertiary care and 2 out of 3 (67%) primary care hospitals have IPC or additional experts for training. Only 1 out of 5 tertiary care hospitals (20%) were recorded in an agreed ratio of healthcare workers to patients while 2 out of 5 (40%) of these hospitals lack staffing need assessment.
Conclusion Overall the sampled public sector (tertiary, secondary and primary) hospitals demonstrated satisfactory IPC level. Challenging areas are the healthcare-associated infection surveillance, monitoring/audit and staffing, bed occupancy overall in all the three categories of hospitals. Periodic training and assessment can facilitate improvement in public sector systems.
- Healthcare quality improvement
- Infection control
- Health Impact Assessment
- Health policy
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information. Not Applicable.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Healthcare-associated infections (HAIs) pose a major threat to patient safety and are a major health challenge. Studies have found that following proper infection prevention and control (IPC) indicators can reduce the incidence of HAIs in hospital environments by up to 70%. The threat of HCAIs in developing nations is 2–20 times higher than in developed countries.
WHAT THIS STUDY ADDS
There are very limited data about IPC systematic implementation in the healthcare facilities of Pakistan. Overall IPC practices are observed deficient and need significant improvement. This is the first study on public sector hospitals of Punjab including primary, secondary and tertiary care level using the IPC Assessment Framework tool.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
The tool enables healthcare policy-makers to acknowledge, assess and promote IPC programmes at the facility level. We need to familiarise our hospitals IPC team with these international standards that can be tailored according to our needs in improving patient care and IPC strategies. The results can be used to develop an action plan for the implementation of the IPC core components at other hospitals, to strengthen existing measures and motivate facilities to intensify efforts where needed.
Introduction
Healthcare-associated infections (HAIs) pose a major threat to patient safety and are a major health challenge. These infections result in increased morbidity and mortality, prolonged hospitalisations, financial drain on healthcare systems and increased antimicrobial resistance (AMR).1 Healthcare facilities are often the epitome of infectious disease outbreaks and are a potential hazard to further spread in the community. Unprepared health systems are unable to withstand the shock of an outbreak, as evident by the West African 2014–2016 Ebola outbreak and the COVID-19 pandemic.2
In the last decade, the WHO has reported a surge in HAIs and AMR, especially in low-income and middle-income countries. This increase is alarming, as these countries have limited resources and inadequate healthcare systems to manage and control these infections effectively.3
Effective implementation of infection control practices is crucial in controlling the transmission of HAIs in settings with high-risk areas in hospitals. For infection prevention measures to be successful, barriers to effective implementation must be identified and overcome.4 WHO and other health organisations have emphasised the importance of implementing effective infection prevention and control (IPC) programmes to curb the spread of these infections. These measures are critical to achieving sustainable development goals, improving patient safety and promoting universal health coverage.5 Healthcare facilities must prioritise IPC and develop protocols that promote the safety and well-being of patients and healthcare workers. The true worldwide burden of HAIs is obscure because of the difficulty in obtaining reliable information.6 Be that as it may, worldwide an expected more than 1.4 million patients at a given time experience the ill effects of HAIs. The threat of HAIs in developing nations is 2–20 times higher than in developed countries.7
IPC is a largely ignored area in hospitals and health facilities across Punjab. A variety of factors including dearth of IPC programmes and resources, fragmented training, education and healthcare structure along with overpopulation contribute to insufficient IPC practices.8 The lack of adequate IPC programmes and resources is a significant challenge that healthcare providers and policy-makers must address.9 The absence of robust IPC measures puts patients and healthcare workers at risk of contracting HAIs, which can lead to significant morbidity and mortality. There is often a lack of coordination between different healthcare facilities, leading to disparities in the quality of care provided to patients. Additionally, the overpopulation in Punjab poses a challenge for healthcare providers to implement effective IPC measures. A cross-sectional study conducted at five hospitals of Islamabad using the WHO Infection Prevention and Control Assessment Framework (IPCAF) reported that the total IPCAF score was less than 200 in all the five hospitals denoting that IPC implementation is deficient and significant improvement is needed.10 The high number of patients seeking care in healthcare facilities in low-income and middle-income countries especially create overcrowding, making it difficult to maintain adequate sanitation and hygiene.11
To prevent HAIs and strengthen IPC structures, WHO recently released (2018) the IPCAF, a questionnaire-like tool designed for assessing IPC structures at the facility level.6 At the facility level, the infection prevention and IPCAF comprises 81 questions across 8 core components (CCs) which address various features of IPC programmes identified by the WHO recommendation.
A hospital’s chief executive/administrator/MS/director in public sector setup is eventually responsible for the delivery of IPC standards. In public sector facilities, there exists a Hospital Infection Control Committee normally chaired by senior clinical specialist. This committee is supported by a team of officers and other staff members including pathologist, administrative staff, infection control nurse (if available) that meet on monthly/quarterly basis and review the IPC practices of the setup. It is imperative that sufficient resources, both financial and human, and administrative support are accessible to the IPC group so that IPC programmes are carried out effectively.12
The study assesses the overall characteristics of IPC standards across selected public sector healthcare facilities using IPCAF as outlined by WHO to assess existing IPC practices and to identify gaps at the facilities level with potential areas of improvement
Methods
Study design
It was a descriptive cross-sectional survey with closed-ended questionnaire, conducted from October 2022 to April 2023. Punjab is the most populated and industrialised province of Pakistan. The public sector healthcare system is organised into primary, secondary and tertiary care facilities. Among the selected public sector hospitals of Punjab, primary/direct care hospitals consists of 100–250 beds which offer services in major specialties such as medicine, surgery, pathology and anaesthesia. Secondary care hospitals are equipped with 251–499 beds and provide additional gynaecology and obstetrics, paediatrics and psychology services. Tertiary care hospitals selected have more than 500 beds and provide sub specialty services such as cardiology, neurology, orthopaedics and other specialised units.
The forces hospitals situated in various cantonments of Punjab were selected for study. We conducted surveys from all the three tiers that is, primary, secondary and tertiary care hospitals to get an idea of overall IPC practices in these hospitals. Sampling unit is hospital. Out of a total of 13 hospitals, two hospitals did not respond so study was conducted in total 11 hospitals; five from tertiary care, three from secondary care and three from primary/direct care hospitals. As we have kept the names of the hospitals anonymous, we designated the following codes to these hospitals. Five tertiary care hospitals (class A) as A1, A2, A3, A4 and A5. Three secondary care hospitals (class B) as B1, B2 and B3 and three primary/direct care hospitals (class C) as C1, C2 and C3.
Study tool
IPCAF (attached in online supplemental annexure A) is a structured, self-administered, closed-formatted questionnaire with an associated scoring system. It is a tool to support the implementation of the WHO guidelines on CCs of IPC programmes at the healthcare facility level. The goal of the framework is to assess the current IPC situation in the hospital setting and identify strengths and gaps that can inform future plans. It is a universal tool that is valid for assessment of IPC standards in any country/facility. The score will be calculated based on eight CCs. The CCs are as follows: (1) IPC programme (CC1), (2) IPC guidelines (CC2), (3) IPC education (CC3), (4) HAI surveillance (CC4), (5) multimodal strategies (CC5), (6) monitoring/audit of IPC practices and feedback (CC6), (7) workload, staffing and bed occupancy (CC7) and (8) environments, materials and equipment for IPC (CC8).
Supplemental material
For every CC, the scores of the individual questions were aggregated. A minimum score of 0 and a maximum score of 100 per CC was assigned. The final IPCAF score was calculated by adding the scores of all eight CCs. The maximum possible total score was 800. Based on the overall score, each hospital was assigned to one of four levels of IPC practice:
Inadequate (0–200 points): Insufficient IPC implementation with significant improvement required.
Basic (201–400 points): Only some functional IPC aspects without adequate implementation with additional improvement needed.
Intermediate (401–600 points): Most IPC aspects are properly implemented.
Advanced (601–800 points): IPC CCs are fully implemented according to WHO recommendations and apposite to the facility needs.
Once the baseline assessment has been made by the IPCAF questionnaire, an online zoom interview was arranged to get inside information from the highest and lowest scoring hospital facility.
Data collection procedure
Hospitals chief executives were approached explaining the purpose of the study and its impact. They were requested to appoint the head of infection control committee or any other officer, a focal person to communicate along with infection control nurse (if available) and who was responsible in consulting other stake holders pertaining to the hospital to complete the essential questionnaire item. A descriptive analysis of each individual CC was performed consequently by a comparative analysis at interfacility and intrafacility level. Each individual healthcare facility further had its average CC and subcomponent scores determined through calculation. To remove any biases and confidentiality purpose names of hospitals are not disclosed. The IPCAF form was sent to the respective hospitals by courier service. The researchers were in contact with the designated focal person of each hospital through telephone to ensure timely submission of the IPCAF form. After completion, all these handwritten filled forms were received again through TCS and data were compiled.
After analysing the scores from the 11 hospitals, highest and lowest scoring hospitals were selected for in-depth interviews from the concerned persons. Since IPCAF scoring system is a standard to assess the existing IPC standards and identify gaps for ongoing improvements, we wanted to assess the understanding and interpretation or any element of personal bias that could affect the transparency of the filled questionnaire.
Statistical analysis
Data entries were completed through MS Access. We analysed the data using IBM SPSS V.25. Due to the scoring system of IPCAF, any unanswered questions were entered as zero. Descriptive statistics presented as frequencies and percentages in tables. We summarised the indicators by frequency, percentage and median with IQR. The data of IPCAF CCs are mostly skewed and do not fit the normal distribution, hence we preferred median with IQR as it is a more robust measure of central tendency. IPCAF scores were compared between same level hospitals as well as different levels hospitals.
Results
Demographic information: 11 public sector hospitals from Punjab were randomly selected. These included tertiary (class A), secondary (class B) and primary/direct care hospitals (class C), respectively, and coding assigned as mentioned in methodology.
Distribution of IPCAF score
The overall median IPCAF score for the participating hospitals was 532.5 (IQR: 292.5–690). Based on CC score distribution, the highest median score (100, IQR: 52.5–100.0) was recorded for IPC Guidelines (CC2), while HAI surveillance (CC4) had the lowest score (median 52.5, IQR: 15–87.5) and the widest variability, and the narrowest for the workload, staffing and bed occupancy (CC7) (table 1).
The hospital with the highest IPCAF scores (770;figures 1 and 2) obtained ‘advanced’ IPC levels whereas the lowest IPCAF score (227.5; figures 1 and 2) managed to reach ‘basic’ IPC levels according to the WHO reference range on IPCAF classification. Both of these are class C, that is, tertiary care hospitals. Four hospitals overall scored advanced category (two from class A and two from class C category). One hospital from class A and two hospitals from class B scored ‘intermediate’ IPC levels. The remaining four hospitals scored ‘basic’ IPC levels (two from class A, one from class B and C each) (figures 1 and 2).
As shown in table 2, overall IPC programme (C1) and IPC guidelines (C2) are well established in most of the hospitals except two tertiary care and one primary care who scored less than 50. All the secondary care hospitals scored more than 70 in these two domains except one. IPC education and training (C3) scores are mostly affected in tertiary care hospitals where three hospitals scored less than 40. The lowest median score was found for HAI surveillance (C4) with widest variability of 52.5 (15, 87.5). Each of the three categories had one hospital where the score was less or equal to 15. More than 50% of primary, secondary and tertiary care hospitals scored above 60 in multimodal strategies (C5). Monitoring and audit of IPC (C6) as well as workload and staffing was not at optimal level as per WHO criteria in all facilities but especially tertiary care facilities. There was good stock and supply management of materials and equipment for IPC in most of the tertiary and primary care facilities (table 2).
Component-based analysis (online supplemental table 3) with key findings of IPCAF assessment in selected tertiary care (class A), secondary care (class B) and primary care (class C) public sector hospitals of Punjab attached in online supplemental file).
IPC programme (CC1)
The calculated median score for IPC programmes was 77.5 (IQR 47.5–92.5). All 11 tertiary, secondary and primary care hospitals have an IPC Programme (100%). Overall, 2 out of 5 tertiary (40%), 1 out of 3 secondary (33%) and 2 out of 3 (67%) primary care hospitals have IPC team/focal person having dedicated time for IPC activities. Regarding microbiological lab support for routine day-to-day use, 4 out of 5 tertiary care hospitals had it and all the 3 hospitals (100%) each from secondary and primary care hospitals had lab capacity to deliver the results reliably (online supplemental table 3).
Supplemental material
IPC guidelines (CC2)
The calculated median score for IPC Programmes was 100(IQR 52.5–100.0). Hand hygiene and transmission-based precautions are again fully followed by all the three categories of hospitals (100%). Only 2 out of 5 tertiary care hospitals (40%) received healthcare provider’s specific training on updated IPC guidelines. Same is for secondary and primary care hospitals where 2 out of 3 hospitals received the training (67% for each category) Regarding disinfection/sterilisation, healthcare worker protection and safety, injection safety 2 out of 3 primary aand secondary care hospitals (67%) followed the protocol. For tertiary care hospitals the percentages are 100, 80 and 100, respectively (online supplemental table 3)
IPC education and training (CC3)
IPC education and training (CC3) varied highly with a median score of 55 (IQR 20.0–80). Lowest number of tertiary, secondary and primary care hospitals were recorded in arrangements for specific IPC training for patients and their family members (20%, 33% and 33%, respectively). Again only 1 out of 5 tertiary care hospitals is offering ongoing development/education for IPC staff (20%). These hospitals also lacked in providing IPC expert or non IPC personnel to lead IPC training (40% each) whereas 2 out of 3 primary and secondary care hospitals (67% each) have this facility.(online supplemental table 3)
Hai surveillance (CC4)
The calculated median score for HAI surveillance was found to be lowest 52.5 (15, 87.5). Only 1 out of 5 (20%) tertiary care hospitals have IT support to conduct surveillance. Similarly more than half of these hospitals lacked trained surveillance professionals, surveillance for infections in vulnerable population, standardised data collection methods or any tailored facility-based plans made accordingly or regular feedback surveillance information. One out of three secondary and primary care hospitals followed device associated infection surveillance, surveillance for local priority epidemic—prone infections, reliable surveillance case definitions, tailored facility-based plans using surveillance data or regular feedback surveillance information (online supplemental table 3).
Multimodal strategies (CC5)
The calculated median score for multimodal strategies was 60(IQR 30.0–75). Only two out of five tertiary care and one out of three primary care hospitals use multimodal strategies in safety climate and culture change. Regarding checklists for these strategies only one out of five tertiary care, one secondary and two primary care hospitals included these (online supplemental table 3).
Monitoring and audit of IPC practices and feedback (CC6)
The calculated median score for monitoring and audit of IPC practices was also quite low 55(IQR 35.0–87.5). Regarding monitoring of hand hygiene compliance, intravascular catheter insertion, wound dressing change, transmission-based precautions, disinfection/sterilisation of medical equipment’s, hand rub and usage of antimicrobial agents, 4 out of 5 (80%) tertiary care hospitals follow compliance while both secondary and primary care hospitals have 100% monitoring. All the three categories of hospitals had lowest number in monitoring of IPC processes in a blame-free culture and assessing safety cultural factors among facility (40%, 33%, 0) (online supplemental table 3).
Workload, staffing and bed occupancy (CC7)
Workload, staffing and bed occupancy also had a calculated median score of 55 (IQR 45–75) with the narrowest variability. Staffing level assessments were found in 3 out of 5 (60%) tertiary care and 2 out of 3 (67%) secondary and primary care hospitals. Lowest no of tertiary care hospitals (20%) were recorded in an agreed ratio of healthcare workers to patients. Overall, 3 out of 5 ((60%) tertiary care hospitals follow staffing need assessments and international standards regarding bed capacity (online supplemental table 3).
Built environment, materials and equipment for IPC (CC8)
CC8 varied lowest among study public sector hospitals with a calculated median score of 77.5 (IQR 47.5–92.5). Overall, three out of five tertiary, two out of three secondary and one out of three primary care hospitals had accessible record of cleaning by the cleaners for floors. Water services, Improved latrines, in-patient functional environmental ventilation, incinerator/alternate technology for waste are available in 4 out 5 (80%) tertiary care and 2 out of 3 secondary care (67%) and almost all 3 primary care hospitals (online supplemental table 3).
Interviews
We identified the highest and lowest scoring hospitals in the baseline assessment through IPCAF tool (shown in figures 2 and 3). Core component wise comparison of these 2 hospitals is shown below in figure 3. An in-depth interview of these two hospitals, that is, C1 and C3 (both primary care hospitals) were conducted. Nominated focal persons who had previously filled up the questionnaire were asked to join so that they were well informed about IPCAF tool. The interviews were conducted through web conference in the presence of supervisor and lasted for about an hour each. After conducting interviews with the highest (C1) and lowest scoring (C3) hospitals some inconsistencies were found in results. Both the hospitals belong to primary/direct care hospital class, meaning both had almost similar infrastructure, manpower and other resources. However, respondents filling up the questionnaire had different level of background knowledge, experience with different perception of the IPCAF tool and its understanding. C1 had resulted in overscoring itself without true understanding of the tool while C3 had underestimated its capabilities. C1 seemed to lack focus and although carrying out most of the tasks, their work lacked objectivity and documentary record. Maybe it was because of fear of bringing bad name to the institution or trying to portray a good image of the facility despite the re assurance about anonymity of the hospital name. C3 was very methodical and clear in his approach indicating the difference in personal approach of data filling.
Overall both the facilities had deficient IPC trained professionals. Furthermore both the facilities lacked documentary record regarding training and surveillance, staffing level requirements and dedicated IPC professionals. Documentation or proper record keeping is an especially neglected area due to overwork and multitasking which needs to be addressed for proper surveillance and action plan. In addition to above all, overburdened healthcare personnel should go through self-care and wellness programmes to release physician’s burn out and improve productivity.
Discussion
Our study has helped to give valuable insights about the existing IPC level of selected public sector hospitals of Punjab. There are only two other studies from Punjab to our knowledge who assessed the state of implementation of key IPC factors using IPCAF tool. One was in tertiary care hospitals of Rawalpindi, Islamabad and other conducted in hospitals of underdeveloped areas of Punjab.10 13 Unlikely and surprisingly the data gathered demonstrated a high level of IPC standards in our public sector hospitals, although Pakistan is considered a lower-middle-income developing country with profound healthcare provision challenges.14 Out of 11, 4 primary, secondary and tertiary care public sector hospitals attained the ‘advanced’ level of IPCAF classification. Besides the differences in the scoring of these selected hospitals, we also observed variation in aggregated scores of the respective IPCAF sections. Although the median for CC1, IPC programme was 77.5 where all hospitals had an IPC Programme but the overall availability of a full-time IPC professional was low in all categories of hospital. Similarly most of the hospitals did not have dedicated time for IPC activities. This lack of priority and budget allocation has been a global pervasive challenge.15 In our study CC2, IPC guidelines were found at high par in meeting the IPCAF SCORE, however, only 40% of tertiary care healthcare workers receive specific training related to new/updated guidelines introduced in the facility. Most of the South Asian countries are deficient in IPC training and education.16 Another study from Bangladesh indicated that healthcare workers lacked quality and sufficient training around IPC, with 85% of participants reporting to get no formal training on infection control.17 Other studies from Ghana, Punjab and even Austria have highlighted these issues.10 18 19 Only one out of five tertiary care hospitals have specific IPC training for patients/family members or ongoing development offered to IPC staff. HAI surveillance showed some serious concerns in our public sector hospitals such as lack of trained professionals for surveillance activities and IT support to conduct surveillance and to make tailored unit-based plans for further improvement. Absence of these parameters hampers the effective HAI Surveillance in middle to low income countries where around half of the reported deaths due to infections can be easily prevented with a well-established surveillance system.20 Although the concept of multimodal strategies is globally gaining its place in medicine, some of these aspects are still comparatively new for us. The lack of interdepartmental coordination and overburdened staff especially in tertiary care hospitals might be contributing to this low adherence.13 Studies from Austria and Germany have highlighted similar findings, although they were in high-income settings and technologically more advanced than us.19 21 Most of the tertiary care hospitals also scored low in workload, staffing and bed occupancy levels (CC7). Four out of five tertiary care hospitals did not have an agreed ratio of healthcare workers to patients. Understaffing and overcrowding have been documented as significant risk factors for HAIs in multiple previous studies.22 23 Public sector hospitals in small cities are generally facing overcrowding and inadequate staffing that require high-level institutional support and planning. The policy-makers need to ensure a standard patient to staff ratio and meeting international standards regarding ward design bed capacity.24 Overall, all the three public sector secondary care hospitals met ‘intermediate’ criteria on IPCAF scoring. The last CC (CC8) assesses the infrastructure and equipment for optimum IPC practices in a healthcare setting. Our data of selected hospitals depicted a satisfactory system of hospitals’ water, electricity, light and mechanical ventilation as more than 70% of the facilities had optimum resources.
In-depth study of the IPCAF CCs generated varied outcomes. Respondents had a different level of firsthand knowledge and perception about IPC practices highlighting that a level playfield is required to finish off this questionnaire. Hence IPC focal person filling the form should be aware of this assessment tool beforehand. This further highlights the need for an Epidemiologist/community health expert to be deployed in all public healthcare facilities with massive patient inflow. Proper training regarding orientation of the national/local standards set for IPC would raise focused and dedicated commitment.
Strengths and limitations
This study enabled us to assess the current IPC situation in the most visited and noticeable public sector hospitals of Punjab including primary, secondary and tertiary care level. Hence, it provided us an opportunity to identify existing gaps A limitation of this study is that the report is based on self-report by a designated IPC focal person assigned by the executive director. Hence assessment methodology can be improved with objective data. This was further affirmed by the in-depth interviews of the highest and lowest scoring hospitals focal person. Both have perceived the IPCAF questionnaire differently which led to such varied scores. Overall, more or less they faced the same issues.
Conclusion
Overall current state of IPC framework and processes in selected Punjab hospitals is at a satisfactory level. The standards of public sector hospitals apparently fulfil most aspects of IPC CCs. According to our data the areas of most needed improvement is CC4 (HAI Surveillance) and then CC7 (workload, staffing and bed occupancy) and CC3 (IPC education and training). As Punjab is the most densely populated province of Pakistan, we need to ensure that the resources are enough to meet the demands of public. Periodic assessment through the IPCAF should be encouraged in all healthcare facilities to monitor changes and develop tailored strategies for promoting IPC.
Data availability statement
All data relevant to the study are included in the article or uploaded as online supplemental information. Not Applicable.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and was approved by 259-AAA-ERC-AFPGMI. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
We are thankful to the designated IPC focal persons from each hospital for their role in our study.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Contributors Conception and design of study: AE/FE. Acquisition and analysis of data: AE/FE/HH. Preparation of manuscript: AE/FE/HH. Critical revision and final approval of manuscript: AE/FE/HH.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.