View Article PDF

Patient involvement in research and medical student training is increasingly recognised as the cornerstone of effective treatment regimens and healthcare service delivery.[[1]] However, recruiting a large and diverse population sample is challenging. Participation rates may differ based on the recruitment method and the type(s) of research and teaching. Patients are usually invited to participate in research projects and medical student training through advertising or directly through healthcare professionals. In addition, previous research has suggested that the public may be more cautious regarding genetic studies than other types of medical research.[[2]] However, it is unknown whether these views are representative of the multi-ethnic New Zealand population living with diabetes in Auckland. Furthermore, there is a lack of data on the characteristics of people who participate in research and medical student training.

The Auckland Diabetes Centre Volunteer Database (ADCVD) was conceived to enable patients to register their interest in being contacted for participation in current and future research projects, the training of medical students, or the co-design phase for planned research. The ADCVD was established as a secure internal database, accessed only by centre administrators and Auckland Diabetes Centre (ADC) researchers. The primary intention behind the ADCVD was to streamline patient recruitment for research projects and medical student training. As part of setting up the ADCVD, we also aimed to determine any differences in patient interest based on demographics and types of research, and to investigate what barriers exist and what motivates or enables patient involvement. It was hypothesised that we would see differences based on ethnicity, and that those from higher levels of deprivation would be less likely to express an interest in being involved in research projects and/or training of medical students. Additionally, it was hypothesised that patients would be less willing to participate in genetic studies compared to other types of studies.

Method

This study was conducted following the ethical standards of the Auckland District Health Board (ADHB) and the Health and Disability Ethics Committee (HDEC 18/NTA/36). Scheduling staff members at the ADC who routinely contact patients to arrange clinical appointments also invited patients to enrol in the ADCVD. This approach was deemed less intimidating, and with less obligation than if the healthcare providers were to approach patients directly. All patients who had attended an appointment at the ADC between July 2018 and July 2019 were contacted via text message (SMS). The SMS was written as: “We have diabetes research studies at the Auckland Diabetes Centre. Text YES to be contacted about this. Text NO if you are not interested”. Patients who answered “Yes” were contacted by a scheduling staff member either by email, phone call or SMS and were asked to complete a survey.

The survey was written in English and was composed of six questions (Appendix 1). Its purpose was to capture data on the types of research studies in which patients were willing to participate, and qualitative data on factors influencing patients’ participation decisions. Patients were also asked if they would be interested in being contacted for medical student training. Qualitative data were coded to generate the main themes using an inductive approach by a single researcher. Patient demographic data such as age, gender, ethnicity, and home address were derived from electronic medical records. Ethnicity, as recorded on clinical records, was then aggregated into the categories listed in Table 1. Deprivation was determined using patients’ home addresses and the NZDep Index, an area-based measure of socio-economic deprivation in New Zealand.[[3]]

Quantitative data were analysed using GraphPad Prism 8.2.1 (California, United States of America). A multivariate logistic regression was performed to assess the relationship between SMS response and the explanatory variables: gender, deprivation index, ethnicity and age. Data were checked for multicollinearity with the Belsley-Kuh-Welsch technique. The heteroskedasticity and normality of residuals were assessed by the White test and the Lilliefors test. The Fisher’s exact test was used to assess the relationship between survey responses and demographic variables. A p-value <0.05 was considered statistically significant.

Results

Enrolment

A total of 2,884 patients with diabetes who attended the ADC between July 2018 and July 2019 were sent an SMS by scheduling staff. Of these, 527 (19%) replied “Yes”, 618 (21%) answered “No”, and the remaining 1,739 (60%) did not respond (NR) (Table 1). Patients who responded “Yes” were entered into the ADCVD and were asked to complete the survey. They were provided with the option of completing the questionnaire by phone, email or an online form provided via an SMS link. The survey was completed by 176 patients (email n=146, phone n=18, SMS n=12), and their answers were entered into the ADCVD (Figure 1).

Figure 1: Flow diagram of patient enrolment into the ADCVD.

View Table 1.

Demographics

When comparing individual ethnic groups, the proportion of NZ Europeans that both responded to the SMS and completed the survey was significantly greater than for Māori (p=0.0251), Pacific (p<0.0001), Indian (p<0.0001), other Asian (p<0.0001), and other ethnicities (p<0.0001) (Figure 2).

Figure 2: The percentage of patients invited to be enrolled in the ADCVD differed from the percentage of patients who agreed and completed the survey by ethnicity.

Age did not influence response rate. The median age of patients who agreed to be contacted about diabetes research and completed the survey was 60.8 years (range of 28–76 years). Women were less likely to respond to the SMS than men [OR=0.69 [0.56, 0.85], p=0.0004] and made up only 28% of patients who completed the survey (Figure 3).

Figure 3: The gender distribution of patients changed across the stages of recruitment.

The proportion of males relative to females increased across the stages of recruitment from the initial invitation to survey completion. NR=no response.

As shown in Table 1, individuals from the least deprived quintile (deciles 1 and 2) made up 12.6% (n=363) of those initially contacted, but 19.9% (n=36) of those who completed the survey. Conversely, individuals in the most deprived quintile (deciles 9 and 10) comprised 27.8% (n=803) of those initially contacted, but 20.5% (n=36) of those who completed the survey.

Acceptability of different forms of research

Patient willingness to be involved in specific areas of research was queried, specifically: 1) Genetic studies using blood or saliva, 2) Other studies using blood samples, 3) Questionnaires or surveys, 4) New medication trials, 5) Weight loss studies. There was no difference in the acceptability of the different forms of research studies amongst patients who completed the survey. On average, 86% of patients expressed a willingness to be involved in each area of research. Contrary to our hypothesis, interest in genetic research was similar to that of other research types. Furthermore, willingness to participate in genetic research did not appear to differ by ethnicity (𝛘[[2]]=8.969, df=5, p=0.11). Overall, 73% of respondents expressed an interest in the design of future studies, whilst 78% of respondents expressed a willingness to be involved in medical student training.

Motivations and challenges

We collected qualitative data about research participation from 92 patients (52% of those surveyed) who provided their views in a free text section of the survey. The key themes and supporting quotes are outlined in Table 2.

View Table 2.

Discussion

We created a database of patients attending the ADC who were willing to participate in future research and medical training opportunities. As part of establishing this database, we investigated whether there were demographic differences amongst those interested in taking part, and whether some forms of research were more acceptable than others. We found an overrepresentation of NZ European respondents and an under-representation of patients from Pacific and Asian ethnicities. Based on these results, we cannot conclude that people of non-NZ European descent were less interested in medical research or training; instead, our efforts to engage with these populations may have been insufficient. All study correspondence was in English, giving rise to potential language barriers. Therefore, future attempts to engage with a multi-ethnic cohort of patients should include multilingual correspondence. Involving community leaders may have also helped recruit non-NZ European patients better. However, different ethnic groups may hold different perspectives on the value of research. It has been previously reported overseas that patients who identify as Chinese are less likely to participate in clinical studies, compared to other ethnic groups, due to barriers such as insufficient information provided during recruitment, language, cultural values, and mistrust of research.[[4]] Further research could explore whether barriers to participation differ by ethnicity amongst our population and how to mitigate these.

We found that patients living in more deprived areas would be less likely to show interest in medical research and training. Again, it is difficult to determine whether the low response rate among patients living in greater deprivation is due to a lack of interest or limited means to respond to our initial contact. The cost to reply to the initial SMS contact, and changes in phone numbers, may have prevented responses. Further, the contact number listed may have been inaccurate or shared with several other family members, making this contact mode unreliable. An opt-out default could have been used instead of the traditional opt-in approach. Previous diabetes research using an opt-out default has reported higher enrolment rates but also higher attrition rates.[[6]] Therefore, this suggests that the opt-in approach may reach motivated individuals who do not represent the target population but are more likely to follow through in such research or training activities.

Other recruitment strategies may have also proved useful. For example, patients could have been approached while sitting in the ADC waiting room. Although this is labour intensive, face-to-face—kanohi ki te kanohi—interactions can foster trust in the researcher and build relationships, thereby facilitating successful recruitment.[[7]] The Scottish Health Research Register (SHARE) has successfully recruited many volunteers interested in taking part in research. Using the Community Health Index (CHI) number, individuals are identified for potential studies using information held in National Health Service data sets such as those from hospital discharges, hospital outpatient attendances and primary care prescribing. Interestingly, face-to-face recruitment in outpatient departments and general practitioner practices was their most successful recruitment method, with around 90% of those approached agreeing to join.[[8]] In this way, collaborating with primary care to recruit those who have intimated an interest in participating in health research may better identify potential participants with diabetes.  

In line with previous research,[[2]] we hypothesised that participating in genetic research would be less favourable than other forms of research among our patients. Instead, we observed a similar level of interest across all types of research. Despite these promising results, our study is subject to limitations. Patients chose to complete the survey; thus, our results are susceptible to self-selection bias. The response rate to the SMS invite was low, and the subgroup of respondents in this study may not be equivalent to the entire target population. Also, qualitative data collection consisted mainly of closed questions for brevity. Therefore, the expectations regarding research or medical student training participation may not have been clearly defined, in terms of additional time commitment or other details. Conducting focus groups or conversational interviews can yield more qualitative data and reach saturation of themes; and in doing so, better understand motivations and challenges for patient involvement.

In conclusion, the ADCVD was created to form a primary contact database for future research and medical student training opportunities. SMS-based recruitment did not result in a representative population attending the ADC. Successful patient volunteer database recruitment and maintenance long term will require more funding for the systematic recruitment of volunteers, organisation, administration and interaction with researchers and clinical educators looking for potential volunteers according to various eligibility criteria.

Future engagement should be tailored to suit different contexts and research topics, and to ensure a broad representation of patient demographics and perspectives.

View Appendix.

Summary

Abstract

Aim

To establish interest in medical research and student training, based on demographics of those attending public-funded diabetes services and types of research.

Method

Patients who attended the Auckland Diabetes Centre (ADC) between July 2018 and July 2019 were invited via text message (SMS) to register their interest in being contacted for future health research projects and medical training. Consenting adults were enrolled in the Auckland Diabetes Centre Volunteer Database (ADCVD) and sent a survey on the acceptability of various types of research and factors influencing participation. Relationships between ADCVD enrolment and other variables were determined using Fisher’s exact test. Qualitative data were coded to generate key themes using an inductive approach.

Results

Of 2,884 patients contacted, 527 were enrolled in the ADCVD (response rate: 18.3%); and of these, 176 completed surveys (response rate: 33.3%). Most respondents were NZ European (n=92, 52.3%), male (n=125, 70.6%), and from the least deprived areas (n=35, 19.9%). The type of research did not affect interest. Motivations to participate centred around a hope to improve their own diabetes and that of future generations.

Conclusion

SMS-based recruitment from a diabetes clinic results in modest interest in participation in teaching and research from predominantly those of NZ European ethnicity and living in areas of least socio-economic deprivation.

Author Information

Ry Yves Tweedie-Cullen: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Audrey Tay: Dietitian, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Yiping Zou: Research Assistant, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Rebecca Brandon: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Ryan Yeu: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand. Stacey Ruru: Research Assistant, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Holly Carmichael: Patient Administration Co-ordinator, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Ole Schmiedel: Clinical Service Director, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Rinki Murphy: Associate Professor, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Consultant Diabetologist, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand.

Acknowledgements

Correspondence

Dr Rinki Murphy: Professor, Department of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Consultant Diabetologist, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. +64 9 923 6313.

Correspondence Email

rinkim@adhb.govt.nz

Competing Interests

Nil.

1) Brett J, Staniszewska S, Mockford C, Herron-Marx S, Hughes J, Tysall C, et al. A systematic review of the impact of patient and public involvement on service users, researchers and communities. The Patient-Patient-Centered Outcomes Research. 2014;7(4):387-95.

2) Matsui K, Kita Y, Ueshima H. Informed consent, participation in, and withdrawal from a population based cohort study involving genetic analysis. J Med Ethics. 2005;31(7):385-92.

3) Atkinson J, Salmond C, Crampton P. NZDep2013 Index of Deprivation. New Zealand: Ministry of Health; 2014.

4) Limkakeng A, Phadtare A, Shah J, Vaghasia M, Wei DY, Shah A, et al. Willingness to participate in clinical trials among patients of Chinese heritage: a meta-synthesis. PloS one. 2013;8(1):e51328.

5) Yu D, Zhao Z, Osuagwu UL, Pickering K, Baker J, Cutfield R, et al. Ethnic differences in mortality and hospital admission rates between M&#x101;ori, Pacific, and European New Zealanders with type 2 diabetes between 1994 and 2018: a retrospective, population-based, longitudinal cohort study. The Lancet Global Health. 2021;9(2):e209-e17.

6) Aysola J, Tahirovic E, Troxel AB, Asch DA, Gangemi K, Hodlofski AT, et al. A Randomized Controlled Trial of Opt-In Versus Opt-Out Enrollment Into a Diabetes Behavioral Intervention. American Journal of Health Promotion. 2018;32(3):745-52.

7) Moyle P. A model for Māori research for Māori practitioners. Aotearoa New Zealand Social Work. 2014;1(26):29-38.

8) McKinstry B, Sullivan FM, Vasishta S, Armstrong R, Hanley J, Haughney J, et al. Cohort profile: the Scottish Research register SHARE. A register of people interested in research participation linked to NHS data sets. BMJ Open. 2017;7(2):e013351.

For the PDF of this article,
contact nzmj@nzma.org.nz

View Article PDF

Patient involvement in research and medical student training is increasingly recognised as the cornerstone of effective treatment regimens and healthcare service delivery.[[1]] However, recruiting a large and diverse population sample is challenging. Participation rates may differ based on the recruitment method and the type(s) of research and teaching. Patients are usually invited to participate in research projects and medical student training through advertising or directly through healthcare professionals. In addition, previous research has suggested that the public may be more cautious regarding genetic studies than other types of medical research.[[2]] However, it is unknown whether these views are representative of the multi-ethnic New Zealand population living with diabetes in Auckland. Furthermore, there is a lack of data on the characteristics of people who participate in research and medical student training.

The Auckland Diabetes Centre Volunteer Database (ADCVD) was conceived to enable patients to register their interest in being contacted for participation in current and future research projects, the training of medical students, or the co-design phase for planned research. The ADCVD was established as a secure internal database, accessed only by centre administrators and Auckland Diabetes Centre (ADC) researchers. The primary intention behind the ADCVD was to streamline patient recruitment for research projects and medical student training. As part of setting up the ADCVD, we also aimed to determine any differences in patient interest based on demographics and types of research, and to investigate what barriers exist and what motivates or enables patient involvement. It was hypothesised that we would see differences based on ethnicity, and that those from higher levels of deprivation would be less likely to express an interest in being involved in research projects and/or training of medical students. Additionally, it was hypothesised that patients would be less willing to participate in genetic studies compared to other types of studies.

Method

This study was conducted following the ethical standards of the Auckland District Health Board (ADHB) and the Health and Disability Ethics Committee (HDEC 18/NTA/36). Scheduling staff members at the ADC who routinely contact patients to arrange clinical appointments also invited patients to enrol in the ADCVD. This approach was deemed less intimidating, and with less obligation than if the healthcare providers were to approach patients directly. All patients who had attended an appointment at the ADC between July 2018 and July 2019 were contacted via text message (SMS). The SMS was written as: “We have diabetes research studies at the Auckland Diabetes Centre. Text YES to be contacted about this. Text NO if you are not interested”. Patients who answered “Yes” were contacted by a scheduling staff member either by email, phone call or SMS and were asked to complete a survey.

The survey was written in English and was composed of six questions (Appendix 1). Its purpose was to capture data on the types of research studies in which patients were willing to participate, and qualitative data on factors influencing patients’ participation decisions. Patients were also asked if they would be interested in being contacted for medical student training. Qualitative data were coded to generate the main themes using an inductive approach by a single researcher. Patient demographic data such as age, gender, ethnicity, and home address were derived from electronic medical records. Ethnicity, as recorded on clinical records, was then aggregated into the categories listed in Table 1. Deprivation was determined using patients’ home addresses and the NZDep Index, an area-based measure of socio-economic deprivation in New Zealand.[[3]]

Quantitative data were analysed using GraphPad Prism 8.2.1 (California, United States of America). A multivariate logistic regression was performed to assess the relationship between SMS response and the explanatory variables: gender, deprivation index, ethnicity and age. Data were checked for multicollinearity with the Belsley-Kuh-Welsch technique. The heteroskedasticity and normality of residuals were assessed by the White test and the Lilliefors test. The Fisher’s exact test was used to assess the relationship between survey responses and demographic variables. A p-value <0.05 was considered statistically significant.

Results

Enrolment

A total of 2,884 patients with diabetes who attended the ADC between July 2018 and July 2019 were sent an SMS by scheduling staff. Of these, 527 (19%) replied “Yes”, 618 (21%) answered “No”, and the remaining 1,739 (60%) did not respond (NR) (Table 1). Patients who responded “Yes” were entered into the ADCVD and were asked to complete the survey. They were provided with the option of completing the questionnaire by phone, email or an online form provided via an SMS link. The survey was completed by 176 patients (email n=146, phone n=18, SMS n=12), and their answers were entered into the ADCVD (Figure 1).

Figure 1: Flow diagram of patient enrolment into the ADCVD.

View Table 1.

Demographics

When comparing individual ethnic groups, the proportion of NZ Europeans that both responded to the SMS and completed the survey was significantly greater than for Māori (p=0.0251), Pacific (p<0.0001), Indian (p<0.0001), other Asian (p<0.0001), and other ethnicities (p<0.0001) (Figure 2).

Figure 2: The percentage of patients invited to be enrolled in the ADCVD differed from the percentage of patients who agreed and completed the survey by ethnicity.

Age did not influence response rate. The median age of patients who agreed to be contacted about diabetes research and completed the survey was 60.8 years (range of 28–76 years). Women were less likely to respond to the SMS than men [OR=0.69 [0.56, 0.85], p=0.0004] and made up only 28% of patients who completed the survey (Figure 3).

Figure 3: The gender distribution of patients changed across the stages of recruitment.

The proportion of males relative to females increased across the stages of recruitment from the initial invitation to survey completion. NR=no response.

As shown in Table 1, individuals from the least deprived quintile (deciles 1 and 2) made up 12.6% (n=363) of those initially contacted, but 19.9% (n=36) of those who completed the survey. Conversely, individuals in the most deprived quintile (deciles 9 and 10) comprised 27.8% (n=803) of those initially contacted, but 20.5% (n=36) of those who completed the survey.

Acceptability of different forms of research

Patient willingness to be involved in specific areas of research was queried, specifically: 1) Genetic studies using blood or saliva, 2) Other studies using blood samples, 3) Questionnaires or surveys, 4) New medication trials, 5) Weight loss studies. There was no difference in the acceptability of the different forms of research studies amongst patients who completed the survey. On average, 86% of patients expressed a willingness to be involved in each area of research. Contrary to our hypothesis, interest in genetic research was similar to that of other research types. Furthermore, willingness to participate in genetic research did not appear to differ by ethnicity (𝛘[[2]]=8.969, df=5, p=0.11). Overall, 73% of respondents expressed an interest in the design of future studies, whilst 78% of respondents expressed a willingness to be involved in medical student training.

Motivations and challenges

We collected qualitative data about research participation from 92 patients (52% of those surveyed) who provided their views in a free text section of the survey. The key themes and supporting quotes are outlined in Table 2.

View Table 2.

Discussion

We created a database of patients attending the ADC who were willing to participate in future research and medical training opportunities. As part of establishing this database, we investigated whether there were demographic differences amongst those interested in taking part, and whether some forms of research were more acceptable than others. We found an overrepresentation of NZ European respondents and an under-representation of patients from Pacific and Asian ethnicities. Based on these results, we cannot conclude that people of non-NZ European descent were less interested in medical research or training; instead, our efforts to engage with these populations may have been insufficient. All study correspondence was in English, giving rise to potential language barriers. Therefore, future attempts to engage with a multi-ethnic cohort of patients should include multilingual correspondence. Involving community leaders may have also helped recruit non-NZ European patients better. However, different ethnic groups may hold different perspectives on the value of research. It has been previously reported overseas that patients who identify as Chinese are less likely to participate in clinical studies, compared to other ethnic groups, due to barriers such as insufficient information provided during recruitment, language, cultural values, and mistrust of research.[[4]] Further research could explore whether barriers to participation differ by ethnicity amongst our population and how to mitigate these.

We found that patients living in more deprived areas would be less likely to show interest in medical research and training. Again, it is difficult to determine whether the low response rate among patients living in greater deprivation is due to a lack of interest or limited means to respond to our initial contact. The cost to reply to the initial SMS contact, and changes in phone numbers, may have prevented responses. Further, the contact number listed may have been inaccurate or shared with several other family members, making this contact mode unreliable. An opt-out default could have been used instead of the traditional opt-in approach. Previous diabetes research using an opt-out default has reported higher enrolment rates but also higher attrition rates.[[6]] Therefore, this suggests that the opt-in approach may reach motivated individuals who do not represent the target population but are more likely to follow through in such research or training activities.

Other recruitment strategies may have also proved useful. For example, patients could have been approached while sitting in the ADC waiting room. Although this is labour intensive, face-to-face—kanohi ki te kanohi—interactions can foster trust in the researcher and build relationships, thereby facilitating successful recruitment.[[7]] The Scottish Health Research Register (SHARE) has successfully recruited many volunteers interested in taking part in research. Using the Community Health Index (CHI) number, individuals are identified for potential studies using information held in National Health Service data sets such as those from hospital discharges, hospital outpatient attendances and primary care prescribing. Interestingly, face-to-face recruitment in outpatient departments and general practitioner practices was their most successful recruitment method, with around 90% of those approached agreeing to join.[[8]] In this way, collaborating with primary care to recruit those who have intimated an interest in participating in health research may better identify potential participants with diabetes.  

In line with previous research,[[2]] we hypothesised that participating in genetic research would be less favourable than other forms of research among our patients. Instead, we observed a similar level of interest across all types of research. Despite these promising results, our study is subject to limitations. Patients chose to complete the survey; thus, our results are susceptible to self-selection bias. The response rate to the SMS invite was low, and the subgroup of respondents in this study may not be equivalent to the entire target population. Also, qualitative data collection consisted mainly of closed questions for brevity. Therefore, the expectations regarding research or medical student training participation may not have been clearly defined, in terms of additional time commitment or other details. Conducting focus groups or conversational interviews can yield more qualitative data and reach saturation of themes; and in doing so, better understand motivations and challenges for patient involvement.

In conclusion, the ADCVD was created to form a primary contact database for future research and medical student training opportunities. SMS-based recruitment did not result in a representative population attending the ADC. Successful patient volunteer database recruitment and maintenance long term will require more funding for the systematic recruitment of volunteers, organisation, administration and interaction with researchers and clinical educators looking for potential volunteers according to various eligibility criteria.

Future engagement should be tailored to suit different contexts and research topics, and to ensure a broad representation of patient demographics and perspectives.

View Appendix.

Summary

Abstract

Aim

To establish interest in medical research and student training, based on demographics of those attending public-funded diabetes services and types of research.

Method

Patients who attended the Auckland Diabetes Centre (ADC) between July 2018 and July 2019 were invited via text message (SMS) to register their interest in being contacted for future health research projects and medical training. Consenting adults were enrolled in the Auckland Diabetes Centre Volunteer Database (ADCVD) and sent a survey on the acceptability of various types of research and factors influencing participation. Relationships between ADCVD enrolment and other variables were determined using Fisher’s exact test. Qualitative data were coded to generate key themes using an inductive approach.

Results

Of 2,884 patients contacted, 527 were enrolled in the ADCVD (response rate: 18.3%); and of these, 176 completed surveys (response rate: 33.3%). Most respondents were NZ European (n=92, 52.3%), male (n=125, 70.6%), and from the least deprived areas (n=35, 19.9%). The type of research did not affect interest. Motivations to participate centred around a hope to improve their own diabetes and that of future generations.

Conclusion

SMS-based recruitment from a diabetes clinic results in modest interest in participation in teaching and research from predominantly those of NZ European ethnicity and living in areas of least socio-economic deprivation.

Author Information

Ry Yves Tweedie-Cullen: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Audrey Tay: Dietitian, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Yiping Zou: Research Assistant, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Rebecca Brandon: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Ryan Yeu: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand. Stacey Ruru: Research Assistant, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Holly Carmichael: Patient Administration Co-ordinator, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Ole Schmiedel: Clinical Service Director, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Rinki Murphy: Associate Professor, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Consultant Diabetologist, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand.

Acknowledgements

Correspondence

Dr Rinki Murphy: Professor, Department of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Consultant Diabetologist, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. +64 9 923 6313.

Correspondence Email

rinkim@adhb.govt.nz

Competing Interests

Nil.

1) Brett J, Staniszewska S, Mockford C, Herron-Marx S, Hughes J, Tysall C, et al. A systematic review of the impact of patient and public involvement on service users, researchers and communities. The Patient-Patient-Centered Outcomes Research. 2014;7(4):387-95.

2) Matsui K, Kita Y, Ueshima H. Informed consent, participation in, and withdrawal from a population based cohort study involving genetic analysis. J Med Ethics. 2005;31(7):385-92.

3) Atkinson J, Salmond C, Crampton P. NZDep2013 Index of Deprivation. New Zealand: Ministry of Health; 2014.

4) Limkakeng A, Phadtare A, Shah J, Vaghasia M, Wei DY, Shah A, et al. Willingness to participate in clinical trials among patients of Chinese heritage: a meta-synthesis. PloS one. 2013;8(1):e51328.

5) Yu D, Zhao Z, Osuagwu UL, Pickering K, Baker J, Cutfield R, et al. Ethnic differences in mortality and hospital admission rates between M&#x101;ori, Pacific, and European New Zealanders with type 2 diabetes between 1994 and 2018: a retrospective, population-based, longitudinal cohort study. The Lancet Global Health. 2021;9(2):e209-e17.

6) Aysola J, Tahirovic E, Troxel AB, Asch DA, Gangemi K, Hodlofski AT, et al. A Randomized Controlled Trial of Opt-In Versus Opt-Out Enrollment Into a Diabetes Behavioral Intervention. American Journal of Health Promotion. 2018;32(3):745-52.

7) Moyle P. A model for Māori research for Māori practitioners. Aotearoa New Zealand Social Work. 2014;1(26):29-38.

8) McKinstry B, Sullivan FM, Vasishta S, Armstrong R, Hanley J, Haughney J, et al. Cohort profile: the Scottish Research register SHARE. A register of people interested in research participation linked to NHS data sets. BMJ Open. 2017;7(2):e013351.

For the PDF of this article,
contact nzmj@nzma.org.nz

View Article PDF

Patient involvement in research and medical student training is increasingly recognised as the cornerstone of effective treatment regimens and healthcare service delivery.[[1]] However, recruiting a large and diverse population sample is challenging. Participation rates may differ based on the recruitment method and the type(s) of research and teaching. Patients are usually invited to participate in research projects and medical student training through advertising or directly through healthcare professionals. In addition, previous research has suggested that the public may be more cautious regarding genetic studies than other types of medical research.[[2]] However, it is unknown whether these views are representative of the multi-ethnic New Zealand population living with diabetes in Auckland. Furthermore, there is a lack of data on the characteristics of people who participate in research and medical student training.

The Auckland Diabetes Centre Volunteer Database (ADCVD) was conceived to enable patients to register their interest in being contacted for participation in current and future research projects, the training of medical students, or the co-design phase for planned research. The ADCVD was established as a secure internal database, accessed only by centre administrators and Auckland Diabetes Centre (ADC) researchers. The primary intention behind the ADCVD was to streamline patient recruitment for research projects and medical student training. As part of setting up the ADCVD, we also aimed to determine any differences in patient interest based on demographics and types of research, and to investigate what barriers exist and what motivates or enables patient involvement. It was hypothesised that we would see differences based on ethnicity, and that those from higher levels of deprivation would be less likely to express an interest in being involved in research projects and/or training of medical students. Additionally, it was hypothesised that patients would be less willing to participate in genetic studies compared to other types of studies.

Method

This study was conducted following the ethical standards of the Auckland District Health Board (ADHB) and the Health and Disability Ethics Committee (HDEC 18/NTA/36). Scheduling staff members at the ADC who routinely contact patients to arrange clinical appointments also invited patients to enrol in the ADCVD. This approach was deemed less intimidating, and with less obligation than if the healthcare providers were to approach patients directly. All patients who had attended an appointment at the ADC between July 2018 and July 2019 were contacted via text message (SMS). The SMS was written as: “We have diabetes research studies at the Auckland Diabetes Centre. Text YES to be contacted about this. Text NO if you are not interested”. Patients who answered “Yes” were contacted by a scheduling staff member either by email, phone call or SMS and were asked to complete a survey.

The survey was written in English and was composed of six questions (Appendix 1). Its purpose was to capture data on the types of research studies in which patients were willing to participate, and qualitative data on factors influencing patients’ participation decisions. Patients were also asked if they would be interested in being contacted for medical student training. Qualitative data were coded to generate the main themes using an inductive approach by a single researcher. Patient demographic data such as age, gender, ethnicity, and home address were derived from electronic medical records. Ethnicity, as recorded on clinical records, was then aggregated into the categories listed in Table 1. Deprivation was determined using patients’ home addresses and the NZDep Index, an area-based measure of socio-economic deprivation in New Zealand.[[3]]

Quantitative data were analysed using GraphPad Prism 8.2.1 (California, United States of America). A multivariate logistic regression was performed to assess the relationship between SMS response and the explanatory variables: gender, deprivation index, ethnicity and age. Data were checked for multicollinearity with the Belsley-Kuh-Welsch technique. The heteroskedasticity and normality of residuals were assessed by the White test and the Lilliefors test. The Fisher’s exact test was used to assess the relationship between survey responses and demographic variables. A p-value <0.05 was considered statistically significant.

Results

Enrolment

A total of 2,884 patients with diabetes who attended the ADC between July 2018 and July 2019 were sent an SMS by scheduling staff. Of these, 527 (19%) replied “Yes”, 618 (21%) answered “No”, and the remaining 1,739 (60%) did not respond (NR) (Table 1). Patients who responded “Yes” were entered into the ADCVD and were asked to complete the survey. They were provided with the option of completing the questionnaire by phone, email or an online form provided via an SMS link. The survey was completed by 176 patients (email n=146, phone n=18, SMS n=12), and their answers were entered into the ADCVD (Figure 1).

Figure 1: Flow diagram of patient enrolment into the ADCVD.

View Table 1.

Demographics

When comparing individual ethnic groups, the proportion of NZ Europeans that both responded to the SMS and completed the survey was significantly greater than for Māori (p=0.0251), Pacific (p<0.0001), Indian (p<0.0001), other Asian (p<0.0001), and other ethnicities (p<0.0001) (Figure 2).

Figure 2: The percentage of patients invited to be enrolled in the ADCVD differed from the percentage of patients who agreed and completed the survey by ethnicity.

Age did not influence response rate. The median age of patients who agreed to be contacted about diabetes research and completed the survey was 60.8 years (range of 28–76 years). Women were less likely to respond to the SMS than men [OR=0.69 [0.56, 0.85], p=0.0004] and made up only 28% of patients who completed the survey (Figure 3).

Figure 3: The gender distribution of patients changed across the stages of recruitment.

The proportion of males relative to females increased across the stages of recruitment from the initial invitation to survey completion. NR=no response.

As shown in Table 1, individuals from the least deprived quintile (deciles 1 and 2) made up 12.6% (n=363) of those initially contacted, but 19.9% (n=36) of those who completed the survey. Conversely, individuals in the most deprived quintile (deciles 9 and 10) comprised 27.8% (n=803) of those initially contacted, but 20.5% (n=36) of those who completed the survey.

Acceptability of different forms of research

Patient willingness to be involved in specific areas of research was queried, specifically: 1) Genetic studies using blood or saliva, 2) Other studies using blood samples, 3) Questionnaires or surveys, 4) New medication trials, 5) Weight loss studies. There was no difference in the acceptability of the different forms of research studies amongst patients who completed the survey. On average, 86% of patients expressed a willingness to be involved in each area of research. Contrary to our hypothesis, interest in genetic research was similar to that of other research types. Furthermore, willingness to participate in genetic research did not appear to differ by ethnicity (𝛘[[2]]=8.969, df=5, p=0.11). Overall, 73% of respondents expressed an interest in the design of future studies, whilst 78% of respondents expressed a willingness to be involved in medical student training.

Motivations and challenges

We collected qualitative data about research participation from 92 patients (52% of those surveyed) who provided their views in a free text section of the survey. The key themes and supporting quotes are outlined in Table 2.

View Table 2.

Discussion

We created a database of patients attending the ADC who were willing to participate in future research and medical training opportunities. As part of establishing this database, we investigated whether there were demographic differences amongst those interested in taking part, and whether some forms of research were more acceptable than others. We found an overrepresentation of NZ European respondents and an under-representation of patients from Pacific and Asian ethnicities. Based on these results, we cannot conclude that people of non-NZ European descent were less interested in medical research or training; instead, our efforts to engage with these populations may have been insufficient. All study correspondence was in English, giving rise to potential language barriers. Therefore, future attempts to engage with a multi-ethnic cohort of patients should include multilingual correspondence. Involving community leaders may have also helped recruit non-NZ European patients better. However, different ethnic groups may hold different perspectives on the value of research. It has been previously reported overseas that patients who identify as Chinese are less likely to participate in clinical studies, compared to other ethnic groups, due to barriers such as insufficient information provided during recruitment, language, cultural values, and mistrust of research.[[4]] Further research could explore whether barriers to participation differ by ethnicity amongst our population and how to mitigate these.

We found that patients living in more deprived areas would be less likely to show interest in medical research and training. Again, it is difficult to determine whether the low response rate among patients living in greater deprivation is due to a lack of interest or limited means to respond to our initial contact. The cost to reply to the initial SMS contact, and changes in phone numbers, may have prevented responses. Further, the contact number listed may have been inaccurate or shared with several other family members, making this contact mode unreliable. An opt-out default could have been used instead of the traditional opt-in approach. Previous diabetes research using an opt-out default has reported higher enrolment rates but also higher attrition rates.[[6]] Therefore, this suggests that the opt-in approach may reach motivated individuals who do not represent the target population but are more likely to follow through in such research or training activities.

Other recruitment strategies may have also proved useful. For example, patients could have been approached while sitting in the ADC waiting room. Although this is labour intensive, face-to-face—kanohi ki te kanohi—interactions can foster trust in the researcher and build relationships, thereby facilitating successful recruitment.[[7]] The Scottish Health Research Register (SHARE) has successfully recruited many volunteers interested in taking part in research. Using the Community Health Index (CHI) number, individuals are identified for potential studies using information held in National Health Service data sets such as those from hospital discharges, hospital outpatient attendances and primary care prescribing. Interestingly, face-to-face recruitment in outpatient departments and general practitioner practices was their most successful recruitment method, with around 90% of those approached agreeing to join.[[8]] In this way, collaborating with primary care to recruit those who have intimated an interest in participating in health research may better identify potential participants with diabetes.  

In line with previous research,[[2]] we hypothesised that participating in genetic research would be less favourable than other forms of research among our patients. Instead, we observed a similar level of interest across all types of research. Despite these promising results, our study is subject to limitations. Patients chose to complete the survey; thus, our results are susceptible to self-selection bias. The response rate to the SMS invite was low, and the subgroup of respondents in this study may not be equivalent to the entire target population. Also, qualitative data collection consisted mainly of closed questions for brevity. Therefore, the expectations regarding research or medical student training participation may not have been clearly defined, in terms of additional time commitment or other details. Conducting focus groups or conversational interviews can yield more qualitative data and reach saturation of themes; and in doing so, better understand motivations and challenges for patient involvement.

In conclusion, the ADCVD was created to form a primary contact database for future research and medical student training opportunities. SMS-based recruitment did not result in a representative population attending the ADC. Successful patient volunteer database recruitment and maintenance long term will require more funding for the systematic recruitment of volunteers, organisation, administration and interaction with researchers and clinical educators looking for potential volunteers according to various eligibility criteria.

Future engagement should be tailored to suit different contexts and research topics, and to ensure a broad representation of patient demographics and perspectives.

View Appendix.

Summary

Abstract

Aim

To establish interest in medical research and student training, based on demographics of those attending public-funded diabetes services and types of research.

Method

Patients who attended the Auckland Diabetes Centre (ADC) between July 2018 and July 2019 were invited via text message (SMS) to register their interest in being contacted for future health research projects and medical training. Consenting adults were enrolled in the Auckland Diabetes Centre Volunteer Database (ADCVD) and sent a survey on the acceptability of various types of research and factors influencing participation. Relationships between ADCVD enrolment and other variables were determined using Fisher’s exact test. Qualitative data were coded to generate key themes using an inductive approach.

Results

Of 2,884 patients contacted, 527 were enrolled in the ADCVD (response rate: 18.3%); and of these, 176 completed surveys (response rate: 33.3%). Most respondents were NZ European (n=92, 52.3%), male (n=125, 70.6%), and from the least deprived areas (n=35, 19.9%). The type of research did not affect interest. Motivations to participate centred around a hope to improve their own diabetes and that of future generations.

Conclusion

SMS-based recruitment from a diabetes clinic results in modest interest in participation in teaching and research from predominantly those of NZ European ethnicity and living in areas of least socio-economic deprivation.

Author Information

Ry Yves Tweedie-Cullen: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Audrey Tay: Dietitian, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Yiping Zou: Research Assistant, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Rebecca Brandon: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Ryan Yeu: Research Fellow, School of Medicine, Faculty of Medical and Health Sciences, The University of Auckland, Auckland, New Zealand. Stacey Ruru: Research Assistant, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand. Holly Carmichael: Patient Administration Co-ordinator, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Ole Schmiedel: Clinical Service Director, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. Rinki Murphy: Associate Professor, School of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Consultant Diabetologist, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand.

Acknowledgements

Correspondence

Dr Rinki Murphy: Professor, Department of Medicine, Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand; Consultant Diabetologist, Auckland Diabetes Centre, Greenlane Clinical Centre, Auckland District Health Board, Auckland, New Zealand. +64 9 923 6313.

Correspondence Email

rinkim@adhb.govt.nz

Competing Interests

Nil.

1) Brett J, Staniszewska S, Mockford C, Herron-Marx S, Hughes J, Tysall C, et al. A systematic review of the impact of patient and public involvement on service users, researchers and communities. The Patient-Patient-Centered Outcomes Research. 2014;7(4):387-95.

2) Matsui K, Kita Y, Ueshima H. Informed consent, participation in, and withdrawal from a population based cohort study involving genetic analysis. J Med Ethics. 2005;31(7):385-92.

3) Atkinson J, Salmond C, Crampton P. NZDep2013 Index of Deprivation. New Zealand: Ministry of Health; 2014.

4) Limkakeng A, Phadtare A, Shah J, Vaghasia M, Wei DY, Shah A, et al. Willingness to participate in clinical trials among patients of Chinese heritage: a meta-synthesis. PloS one. 2013;8(1):e51328.

5) Yu D, Zhao Z, Osuagwu UL, Pickering K, Baker J, Cutfield R, et al. Ethnic differences in mortality and hospital admission rates between M&#x101;ori, Pacific, and European New Zealanders with type 2 diabetes between 1994 and 2018: a retrospective, population-based, longitudinal cohort study. The Lancet Global Health. 2021;9(2):e209-e17.

6) Aysola J, Tahirovic E, Troxel AB, Asch DA, Gangemi K, Hodlofski AT, et al. A Randomized Controlled Trial of Opt-In Versus Opt-Out Enrollment Into a Diabetes Behavioral Intervention. American Journal of Health Promotion. 2018;32(3):745-52.

7) Moyle P. A model for Māori research for Māori practitioners. Aotearoa New Zealand Social Work. 2014;1(26):29-38.

8) McKinstry B, Sullivan FM, Vasishta S, Armstrong R, Hanley J, Haughney J, et al. Cohort profile: the Scottish Research register SHARE. A register of people interested in research participation linked to NHS data sets. BMJ Open. 2017;7(2):e013351.

Contact diana@nzma.org.nz
for the PDF of this article

Subscriber Content

The full contents of this pages only available to subscribers.
Login, subscribe or email nzmj@nzma.org.nz to purchase this article.

LOGINSUBSCRIBE