Table 4 Quality evaluation of included studies using the Mixed Methods Appraisal Tool, 2018 version⁵⁶

1.1

1.2

1.3

1.4

1.5

4.1

4.2

4.3

4.4

4.5

5.1

5.2

5.3

5.4

5.5

Abdolkhani et al. ⁸

Y

Y

C

C

C

Findings not well backed by quotes; small sample size; sampling bias; response bias risk (personal contacts)

Adler-Milstein & Nong³³

Y

Y

Y

Y

Y

Small sample size; authors acknowledge potential confusion of PGHD, PRO and Remote monitoring by participants which might impact the quality of the responses

Andrews et al. ²⁹

Y

N

C

N

Y

Large sample size; risk of sampling bias towards people with positive view on technology; completion rate not reported

Austin et al. ⁹

Y

Y

Y

Y

Y

Findings well supported by quotes; small number of clinicians

Berkowitz et al. ¹⁰

Y

Y

Y

Y

Y

Findings well supported by quotes; risk of selection bias towards participants with positive view on technology

Bietz et al. ²³

Y

Y

C

N

C

Y

C

Y

N

Y

Y

N

C

C

N

Findings are not backed by sufficient quotes which make it difficult to assess the quality of the interpretation and integration with quantitative findings; risk of nonresponse bias (response rate is not reported)

Bruno et al. ³⁴

Y

N

C

C

Y

Limitations in sample size and representativeness due to low response rate (40.8%), female gender bias in individuals and care giver group, sampling bias towards people with technology affinity (online portals were used for recruitment); only limited explanation on questionnaire validation

Cohen et al. ¹¹

Y

Y

Y

Y

Y

Findings are well supported by quotes and compared across the five different studies

Gabriels & Moerenhout⁴⁰

Y

Y

Y

Y

Y

Findings are well supported by quotes; study provides detailed information on interview guide development and data analysis; small sample size

Haase et al. ²⁴

Y

Y

Y

Y

Y

Findings are well supported by quotes; method section of paper already contains results; results section not clearly labelled; risk of response bias (28,75% response rate)

Huh et al. ³⁵

Y

Y

Y

Y

Y

Findings are well supported by quotes, low number of HCP participants, sampling bias for older adults (recruited from a community with high overall education level); only one author performed the analysis

Jacomet et al. ¹²

Y

C

C

N

Y

Information on physicians (e.g., professions, years of experience) missing, nonresponse bias for HIV patients (response rate 59%, gender difference in responder vs non-responder); response rate for HCP not specified, information on questionnaire validation missing

Karduck & Chapman-Novakofski¹³

Y

C

Y

N

Y

Large sample size; representativeness of sample unclear, e.g. almost all participants are female but authors don’t describe the expected demographic variables of the target group; despite overall high response rate of 81% there is a risk of non-response bias (difference between responder and non-responder not defined)

Kelley et al. ³⁶

Y

Y

C

Y

Y

Findings well supported by quotes; student survey questionnaire without details on validation

Keogh et al. ¹⁴

Y

Y

Y

Y

Y

Findings are well supported by quotes; divers sample; potential selection bias (participants might have favourable view on wearables because of project involvement)

Kessel et al. ²⁵

Y

C

Y

N

Y

Low participation (59.1%) and completion rate (37.2%), risk of nonresponse bias

Kim et al. ³²

Y

Y

Y

Y

Y

Y

N

N

C

Y

Y

Y

Y

C

N

Small number of participants, bias in sample (only young clinicians participated), non-response bias undiscussed, questionnaire not piloted

Kong et al. ¹⁵

Y

C

C

N

C

Very low response rate (12.9%), representativeness of sample questionable (target population not specified, low response rate), no information about questionnaire validation, risk of non-response bias

Lavallee et al. ¹⁶

Y

Y

Y

C

C

Findings are not well backed up by enough quotes; sampling bias through purposive sampling

Nguyen et al. ²⁶

Y

Y

Y

Y

Y

Findings were well supported by quotes; small sample size; risk of sampling bias through self-selection of participants; low response rate (10%)

Nundy et al. ³⁷

Y

Y

Y

Y

Y

Y

N

C

N

Y

Y

Y

Y

Y

N

Small sample size and drawn from only one medical centre and potential gender bias (75% female), questionnaire validation not described, risk of non-response bias (31 providers contacted but only 11 interviews completed)

Osborne et al. ¹⁷

Y

Y

Y

C

C

Findings for therapist focus group not backed up by quotes; limited sample size; sampling bias (recruitment of patients from one support group and HCPs from one single clinic)

Ostherr et al. ²⁷

Y

Y

Y

C

C

Findings are not supported by enough quotes; risk of sampling bias (even though the response rate for general public participants was 80%); study miss to discuss limitations

Reading et al. ³⁸

Y

Y

Y

Y

Y

Small HCP sample size; patient sampling bias (predominantly male, middle- to older-age, and moderately to extremely comfortable with technology)

Saleem et al. ²⁸

Y

Y

Y

Y

Y

Risk of sampling bias for already engaged veteran patients and nonresponse bias; frequency of occurrence reported but findings could have been backed up with more original quotes

Sanger et al. ¹⁸

Y

Y

Y

Y

Y

Findings well supported by quotes; risk of sampling bias (participants from one healthcare system), no representation of dark coloured skin patient participants, limited sample size

Sarradon-Eck etal.¹⁹

Y

Y

Y

Y

Y

Risk of nonresponse bias for interviews (86.7% nonresponse rate for purposive sampling, snowball sampling not reported), sample overrepresents GPs in training who might be more interested in mHealth through their teaching activities (sampling bias).

Volpato et al. ³⁰

Y

C

Y

C

C

Mind-maps are innovative but limited for in-depth analysis and potentially inferior to interviews and other qualitative methods; risk of sampling bias

Watt et al. ³¹

Y

Y

Y

Y

Y

Findings well supported by quotes; limited sample size; risk of sampling bias and response bias as some interviews were personal contacts

Wendrich & Krabbenborg²⁰

Y

Y

Y

Y

Y

Findings well supported by quotes; limited sample size; risk of response bias (HCPs might be inclined to report positive views about PGHD as their institutions participate in a pilot study)

West et al. ³⁹

Y

Y

Y

Y

Y

Findings of literature review and interviews were integrated; findings well supported by quotes; limited sample size; risk of sampling bias

Wu et al. ²¹

Y

Y

Y

Y

Y

Findings are well supported by quotes; interview and app analyses integrated; limited sample size, potentially sampling bias through convenience sampling

Zhu et al. ²²

Y

Y

Y

Y

Y

Findings are well supported by quotes; inclusion and exclusion criteria clearly defined, small sample size