Patient-centred care measurement in sub-Saharan Africa: a scoping review of the evidence (1990–2024)

Abstract

The aim of this scoping review was to explore how patient-centred care (PCC) is being measured in healthcare settings in sub-Saharan Africa (SSA) and to examine the psychometric performance of reported measurement instruments for PCC. Medline, Web of Science, EMBASE and Global Health databases were searched for articles published from 1990 to 22 September 2024. Search was updated on 30th December 2024. The keywords ‘patient-centred care’, ‘patient experience’, ‘measurement’ and ‘psychometric property’ were used. Studies were included if they reported on the development, validation, evaluation or psychometric properties of tools used for measuring PCC. Of the 302 articles retrieved, 36, including 31 unique instruments, met the inclusion criteria. Eight instruments were locally developed in SSA. A total of 26 studies were conducted either in South Africa, Nigeria or Ethiopia. Twenty-two studies (60%) were conducted in hospital settings. Psychometric evaluation was either incomplete or not done at all. Locally developed instruments had better psychometric performance compared with instruments developed in the Global North. Although various instruments have been used to measure PCC in SSA, most of them only measure its subcomponents in specific patient populations. Comprehensive measures of PCC, developed and evaluated in SSA for psychometric properties, are needed.

Background

In 2008, the WHO suggested four major reforms to the way healthcare should be delivered.¹ These consisted of public policy, leadership, universal coverage and service delivery reforms. The service delivery reform focused on making services more people-centred with a strong consideration for people’s expectations and needs. This is essential to ensure quality of care and improved health outcomes as improving access to services alone is not a panacea to poor health outcomes, especially in low-resource settings.² For example, despite the increased access to maternity services, with many mothers in sub-Saharan Africa (SSA) now giving birth in the hands of a skilled birth attendant, the burden of complications and maternal deaths have not significantly reduced as expected.³ As a result, governments, institutions, policymakers, clinicians and quality improvement specialists have focused on implementing and measuring progress towards a more patient-centred service delivery. The shift to a more people-centred and community-led delivery of equitable healthcare reportedly led to >18 000 children’s lives being saved per day globally in 2006.¹

Therefore, several instruments have been developed to monitor and measure progress towards a patient-centred health system.⁴ These instruments were developed in high-income settings, and either measured patient-centred care (PCC) as a whole, or more commonly, one of its subcomponents.^4,5 Measuring patient experience and evaluation of PCC became national policies in the early and mid-2000s in both the USA and UK.^6–9 PCC is also measured to evaluate the benefits of new treatment initiatives.¹⁰ It is preferred to other measures, such as measures of patient satisfaction because PCC captures the wider process of and experience with care, and can therefore result in actionable feedback from patients.^11–14 SSA lags behind in the measurement of PCC and patient experience.¹⁵ Currently, only South Africa conducts routine national patient experience surveys, which started in 2017.¹⁶

How has PCC been measured globally?

PCC refers to care that is holistic and responsive to patients’ needs.² The subdomains include respect for patients’ preferences, coordination and integration of care, information and education, physical comfort, emotional support, involvement of family and friends, continuity and transition and access to care.¹⁷ Compared with patient experience and satisfaction questionnaires that are outcome measures, PCC measures the actual process of receiving care, thus making it a preferable measure.¹² Despite this, existing scales measure satisfaction, patient experience and subdomains of PCC when assessing PCC.^12,15

PCC is a complex and dynamic concept that has required various approaches to measurement.¹⁸ Some of the approaches include clinician interviews¹⁹; direct patient interviews and focus group discussions²⁰; observation of clinical encounters and video recordings of patient–provider interactions²¹; examination of patients’ records or routinely collected data²²; and surveys using questionnaires that are either interviewer- or self-administered, posted or online and the use of feedback kiosks for real-time feedback by patients to health facilities.^13,23 Measuring PCC from the patient’s perspective has been proposed as the best way because patients are the ‘experts’ of their care.^4,7,24 In SSA, Afulani and colleagues used a combination of literature reviews and patient interviews to develop and validate a tool for measuring patient-centred maternity care.²⁵

Globally, different tools have been used to measure PCC. A rapid review by the Health Foundation (UK) found 160 measurement tools for PCC.⁴ The most commonly used tools were the Individualised Care Scale,²⁶ Measure of Processes of Care (MPOC)²⁷ and the Person-Centred Care Assessment Tool.²⁸ These three tools measure the holistic concept of PCC compared with, for example, the Jefferson Scale of Physician Empathy, Consultation and Relational Empathy Scale or patient activation measure, which focus on selected components. However, an earlier review by Edvardsson and Innes found that most tools were of poor quality and most lacked patients’ perspectives.²⁹

Another systematic review found that of 88 patient-reported experience measures identified, only two were developed in SSA,³⁰ and psychometric tests were not completely reported for most instruments.^30,31 However, this review excluded instruments that measured important dimensions of PCC such as empathy and patient participation; it also excluded measures involving specific treatment or interventions such as anaesthesia and pharmaceutical dispensation, which are key in patients’ experience of care. In addition, all instruments with unpublished validity and reliability results were also excluded.³⁰ This scoping review was therefore designed to be more inclusive, taking into account patient experiences, and to explore how PCC is being measured in healthcare settings in SSA; and to describe the psychometric performance of reported measurement tools, necessary for evaluating the instrument’s usefulness.³² Although standard guidance for assessing the quality of measurement instruments exists,^33,34 this was not strictly followed in this review as the focus was on description rather than on primarily assessing the quality of available instruments. The guiding questions for the review included: How is PCC being measured in healthcare settings within SSA? What is the (psychometric) performance of the measurement tool(s) being used?

Methods

Search strategy

Medline, Web of Science, EMBASE and Global Health databases were searched for articles published from 1990 to 22 September 2024, except for the Global Health database, in which the search included articles from 1973 to 2024, as predetermined in the database. We used the keywords ‘patient-centred care’, ‘patient experience’, ‘individualised care’, ‘measurement’, ‘psychometric property’ and ‘sub-Saharan Africa’ to inform the search strategy. The full search strategy for all databases is shown in Appendix 1. All searches were updated on 30 December 2024. All updates were checked against the eligibility criteria before merging with already included articles.

Inclusion and exclusion criteria

Studies were included if they reported on the development, validation, evaluation or psychometric properties of tools or questionnaires used for measuring PCC or its related subcomponents or domains or dimensions in healthcare settings within SSA. These domains or dimensions were informed by a prior systematic review conducted by the same research team.³⁵ An instrument was referred to as comprehensive if it measured all these components. Examples of these domains include trust, communication and shared decision-making. We considered community, primary, secondary and tertiary care centres. Compared with the prior systematic review, this paper highlights the psychometric performance of the tools used for measuring PCC in SSA. Studies conducted in English and employing either qualitative or quantitative study designs were included, whereas study protocols, editorials and review articles were excluded.

Study selection and data extraction

PKO retrieved articles and screened titles and abstracts to identify articles for full-text review. The list of the included articles was discussed with coauthors and data were extracted into a summary table, which was cross-checked by CP and JG. Data were extracted for authors and year of publication, country of origin, study aim, study design, population/sample, measurement tool, dimension of PCC measured and psychometric properties reported, if available. Descriptive summaries and charts/tables were used to handle and summarise the retrieved studies. The methodological quality of the included studies was not assessed. See Table 1 for a summary of the characteristics of included studies.

Table 1.

Characteristics of studies included in the review

Authors, year	Country	Aim/main objective	Design + approaches	Population + sample size	Component of PCC measured	Measurement tool	Psychometric properties reported	Comment
Henbest and Fehrsen, 1992	South Africa	Validity and reliability of patient centredness in the consultation measure	Prospective cohort. Consultation audiotaped and assessed by independent raters for its patient-centredness	72 patients, 5 medical doctors and 3 nurses in primary care setting	Holistic	Patient-centredness in the consultation	Reliability: inter-rater r= 0.95, 0.88, 0.87. Found to be sensitive and practical	Mostly used in primary healthcare. Measures mainly communication. Raters were neither patients nor practitioners
Ulys et al., 1997	South Africa	To measure consumer expectations in psychiatric units	Cross-sectional survey using a locally (researcher-) developed instrument	145 psychiatric patients	Patient satisfaction	Researcher-developed instrument based on the GHAA Consumer Satisfaction Survey	Inter-rater reliability=0.9372	This instrument and its manual were published first in 1988, and a revised edition in 1990 (Davies and Ware, 1991). The following are assessed: Access: arranging for and getting care Finances Technical quality Communication Staff attitudes
Westaway et al., 1998	South Africa	Develop an instrument to measure satisfaction with family planning services in Gauteng, South Africa	Cross- sectional: interviews	Participants (244) receiving family planning services in Gauteng, South Africa	Satisfaction: interpersonal and organisational dimensions	Locally developed tool to measure satisfaction with family planning services	Coefficient alpha for the 17-item scale was 0.76	Researchers conducted factor analysis that helped reduce the scale from 20 to 17 items
Olusina et al., 2002	Nigeria	To assess patients’ satisfaction with ward experiences	Cross-sectional, interviewer-administered 16-item questionnaire—Likert scale	118 psychiatric in-patients	Satisfaction	PACQ	Cronbach alpha 0.79. Face and content validity also assessed	Levels of satisfaction for each item were graded as follows: dissatisfaction (<50% of respondents positively appreciated it), bare satisfaction (50–65%), moderate satisfaction (66–74%) and highest satisfaction (≥75%)
Westaway et al., 2002	South Africa	To evaluate the quality of healthcare of black patients with diabetes	Cross-sectional	263 black outpatients with diabetes	Satisfaction	Researcher-developed 25-item questionnaire	Validity (multitrait tests) The reliability coefficients were 0.98 (interpersonal), 0.85 (logistics), 0.82 (technical)	This 25-item patient satisfaction scale was developed and tested for evaluating the quality of healthcare for black outpatients with diabetes
Kolawole et al., 2004	Nigeria	To test the reliability and validity of the DTSQ	Cross-sectional, questionnaire. Both self-administered and interviewer- administered	83 patients with diabetes	Satisfaction	DTSQ	Cronbach alpha=0.74, inter-item correlations=0.22–0.79, item-total correlations=0.39–0.78	Instrument performance similar to previous findings from other investigators. Authors noted high satisfaction scores despite worse/poor treatment conditions
Saloojee et al., 2009	South Africa	Validity and reliability of the MPOC	Interviewer-administered questionnaire	263 caregivers of children with a diagnosis of cerebral palsy	Holistic	MPOC. Modified version of the MPOC called MPOC-22SA	Cronbach’s alpha ranged from 0.24 to 0.66 Intraclass correlation coefficient ranged from 0.51 to 0.61	MPOC was neither reliable nor valid in the South African setting. Performance improved after factor analysis with loading producing the MPOC-8SA
Njilele et al., 2012	Nigeria	To develop and validate a questionnaire for assessing HIV-infected patients’ satisfaction with pharmaceutical services in HIV/AIDS clinics	Interviewer administered5-point Likert scale	400 participants (HIV+ patients)	Satisfaction	Researcher developed 20-item questionnaire. Developed from literature, piloted at a general hospital and validated at a tertiary hospital	Convergent and discriminant validity acceptable. Reliability was 0.85	Tool for measuring satisfaction. Disease-specific HIV patients, pharmaceutical services. Single centre
Afolabi et al., 2012	Nigeria	To develop a patient satisfaction scale to assess the quality of pharmacy services	5-point Likert scale. Interviewer- administered	506 outpatients using three hospital pharmacies	Developed from literature and other scores	35-item PSS. Initially piloted on 30 patients, which led to the reduction of items to 25!	The collected data yielded a KMO measure of sampling adequacy, value of 0.80 that was an indication that the collected data were suitable for factor analysis. Cronbach alpha improved from 0.67 (35-item) to 0.71. Also reported Spearman and Guttman split half coefficients of reliability	The items were subjected to expert (four of them) review for professional judgement on ambiguity, relevance and sentence structure. A previous satisfaction scale was separately developed in Nigeria
Agu et al., 2014	Nigeria	Determine satisfaction of HIV+ patients with pharmaceutical services	Exit interviews (researcher-administered questionnaire). Cross-sectional design	2700 patients with HIV from 17 study sites	Satisfaction	5-point Likert scale developed specifically for this study	None	Satisfaction influenced by education level, not age, gender, marital status
Woldehaimanot et al., 2014	Ethiopia	Experience and satisfaction with postop pain management	Cross-sectional, interviewer-administered questionnaire	252 postop patients	Satisfaction	Modified American Pain Society Patient Outcome questionnaire	Cronbach alpha was 0.78, PCA, KMO and Bartlett’s tests were used to assess for appropriateness of data	3 components assessed separately: pain intensity, pain interference and beliefs about pain. The last items of the questionnaire assessed satisfaction. High satisfaction despite inadequate pain management
Khamis et al., 201487	Tanzania	To determine satisfaction with quality of care	Cross-sectional design. Interviewer-administered. Exit questionnaires	422 patients	Satisfaction	SERVQUAL questionnaire	PCA, KMO and Bartlett’s tests, scree plot/eigen values	Did not assess the patient–doctor interactions and may not reflect the overall quality of care at the hospital. Strength: used the Donabedian model
Okoye et al., 2014	Nigeria	To determine satisfaction of HIV patients with pharmaceutical services. To revalidate the PSPS questionnaire	Cross-sectional design, multistage sampling technique. Self-administered questionnaire after using the pharmacy services	1637 patients with HIV from six HIV clinics in southeast Nigeria	Satisfaction	PSPS questionnaire	Overall Cronbach alpha was 0.85. PCA. Construct validity using Multitrait-Multimethod Matrix by Campbell and Fiske, 1959	Confirmed validation by initial developers of the PSPS. Weakness: response bias and acquiescence. Strength: homogenous population, large sample size
Aloba et al., 2014	Nigeria	To evaluate the psychometric properties of the TPS in Nigeria	Cross-sectional, interviewer-administered scale	223 psychiatric OPD patients at a university hospital	Trust	TPS	Cronbach’s alpha of 0.68. PCA with varimax rotation and Kaiser normalization: loading on two factors. Correlational analysis to assess construct validity	Also identified factors associated with trust in the psychiatrist. Generally weak psychometric properties of the TPS in Nigeria. Mean trust score relatively high (75.74)
Myers B et al., 2015	South Africa	Development and psychometric validation of a tool to measure perceived quality of addiction treatment services	Cross-sectional. Self-administered questionnaire	364 patients	Holistic (person-centred services)	SAATSA	Model indices had acceptable values (χ2/df ratio=2.21; CFI=0.98; RMSEA=0.06, 90 % CI 0.04 to 0.09)	5 domains. Person-centred services was one of five domains being assessed
Sheferaw et al., 2016	Ethiopia	Development and validation of a tool to measure RMC	Cross-sectional, mixed approach (both qualitative and quantitative). Researcher-administered/interview to mothers in puerperium	509 participants	Respectful care (dignity and respect)	15-item tool newly generated by researchers	Cronbach 0.859; KMO + Bartlett’s tests; PCA, validity (content, concurrent, construct) assessments	Rigorously followed steps for scale development. Specific name of the scale not given
Siddharthan et al., 2016	Uganda	To examine the feasibility of implementation of a validated patient-centred education tool among patients with heart failure in Uganda	Prospective cohort, 3 mo follow-up; interviewer-administered	105 patients	Patient activation/engagement	PAM-13	Not reported	PAM developed for chronic disease. PCC was measured using PAM before and after a patient-centred intervention
Matston et al., 2017	Ethiopia	To explore the psychometric properties, validity and reliability of a locally relevant measure of service satisfaction	Mixed methods. Interviewer-administered	400 participants	Satisfaction	MHSSS	Bartlett’s test of sphericity and KMO test, EFA, item-test correlation; inter-item correlation and Cronbach’s alpha (internal consistency). Internal consistency and stability (test-retest) were tested	No CFA done. Rigorous scale development steps
Dullie et al., 2018	Malawi	Develop and validate (psychometric properties of) an instrument to measure PHC performance from patients’ perspective	Cross-sectional. Interviewer-administered	631 patients from one district	Satisfaction	PCAT-Mw	Cronbach ranged from 0.66 to 0.91, a satisfactory goodness of fit model was achieved (GFI=0.90, CFI=0.91, RMSEA=0.05)	Modified the ZA-PCAT which was developed from the original American PCAT. Very bulky, looks at nearly 114 items. Assessment of primary care, not directly PCC. Rigorous process of cultural validation was done
Smiley et al., 2019	Ghana	To evaluate staff and patients’ perceptions of safety and quality. Patients’ care experience	Cross-sectional. Interviewer- administered	15 inpatients and staff members	Satisfaction	HCAHPS and SAQ	Cronbach reported for domains of SAQ but not for HCAHPS	Very small sample sizes: 31 for SAQ and 15 for HCAHPS
Eksteen et al., 2019	South Africa	Determine validity and reliability of the MISS	Interviewer- administered. South African primary care setting	150 patients being seen by either doctors or nurses	Satisfaction	MISS	Reliability of 0.899 overall but low in the subscales (ranging from 0.092 to 0.538)	Reliability of communication comfort and compliance intent were low (unreliable). Only reliability and content validity assessed. No factor analysis. Scale was generally unreliable in the South African setting
Archer et al., 2019	South Africa	Reliability and validity of the JSE	Cross- sectional. Self-administered	206 third-year medical students	Empathy	JSE-S	Reliability of 0.81	Conducted with medical students
Baynes et al., 2019	Tanzania	Experience with PAC	Mixed methods	412 women post-abortion	Satisfaction with quality	Satisfaction scale—researcher developed	Alpha of 0.81	Women were less likely to report satisfaction with care at referral facilities owing primarily to inadequate counselling
Mukamba et al., 2020	Zambia	Satisfaction with HIV services	Cross- sectional, interviewer-administered	442 patients with HIV	Satisfaction with care	Adapted the Adult Primary care Questionnaire	Alpha 0.93; EFA—construct and criterion validity assessed	Satisfaction with care promotes re-engagement with HIV care
Abboah-Offei et al., 2020	Ghana	Feasibility of a community intervention to improve PCC among PLWHIV. Used a cRCT design	Cross-sectional, interviewer-administered. Part of a cRCT	60 patients, 30 in each arm	Patient-centredness (holistic)	CARE and PEQ measures	No psychometric analyses were done	Assessed the feasibility of a novel care-enhanced intervention to improve PCC. PCC was measured using the CARE measure and the PEQ
Sudhinaraset et al., 2020	Kenya	Develop and validate a tool to measure PCAC	Cross- sectional, interviewer- administered	353 women	Person-centredness (holistic)	26-item newly developed tool	Alpha 0.81; performed EFA, construct, content and criterion validity	Followed admirable steps for tool/instrument development
Kassa et al., 2021	Ethiopia	Perception of quality of pharmaceutical services in Tigray region	Cross-sectional, face-to-face interviews	793 patients	Satisfaction	SERVQUAL instrument	Alpha 0.877	Strength: large sample size. Weakness: pharmaceutical services only, outpatients
Ogbuabor et al., 2021	Nigeria	To validate the PCMC scale	Cross- sectional survey. Interviewer-administered	450 mothers	PCMC (holistic)	PCMC scale	The goodness of fit χ2 statistic was 1150.305 (ρ<0.001).	8 items deleted following EFA due to low communalities. PCMC scores correlated strongly with women’s overall satisfaction with quality of maternity care (r 0.910, ρ<0.001) indicating high criterion validity
Clarke-Deelder et al 2022	Tanzania	To develop and validate a tool to measure user experience among caregivers of sick children	Cross-sectional survey. Interviewer-administered (exit interviews)	1085 caregivers from 75 facilities	Care experience	Researcher developed 8-item scale	Alpha 0.7; content validity; construct validity—correlation with overall assessments	8-item composite score for user experience. Score quantified as 0 to 1 converting the scale from categorical to interval, with equal distances
Kassaw et al., 2022	Ethiopia	Measure satisfaction with mental health services and its associated factors	Cross-sectional. Face-to-face interviews	409 participants	Satisfaction	MHSSS	Mean satisfaction score was 64/96. Psychometric assessment not done. Relied on previous reliability of 0.84 reported by Mayston et al., 2017, in Ethiopia	Tool was pretested in 5% of participants and yielded an alpha of 0.85
Ratanjee-Vanmali et al., 2020	South Africa	Satisfaction with a hybrid model of hearing health service delivery	Cross-sectional. Online survey	31 out of 46 patients completed the survey. Required patients to have internet access and digital skills	Satisfaction	SAPS	Psychometric properties not reported	Online survey. Small sample size (31). Part of evaluation of a model of health service delivery. NPS also calculated
Christoffels et al., 2018	South Africa	Assessing the consultation process in primary care clinics	Audio-recorded consultation	45 primary care practitioners (doctors and nurses)	Communication	Stellenbosch University Observation Tool (based on the Calgary–Cambridge guide to consultation)	Not assessed	Consultation was audio-recorded and later assessed by independent external observers (other than practitioners or patients themselves). Assessment based on behaviours of practitioners. Risk of Hawthorne effect when observing or recording participants
Abate et al., 2023	Ethiopia	The level of perceived compassionate care	Cross- sectional, interviewer- administered	423 psychiatric patients attending outpatient clinics of two hospitals in Ethiopia	Compassion and shared decision-making	12-item SCCCS and the SDM-Q-9	Not reported	Examined perceived compassionate care among psychiatric patients. SDM-Q-9 was used for criterion validity
Kagura et al., 2023	South Africa	Chronic disease patients’ levels of satisfaction with care in Johannesburg, South Africa	Cross-sectional, interviewer-administered 22-item newly developed questionnaire	2429 patients with chronic disease at 80 primary healthcare facilities	Patient satisfaction	Newly developed 22-item tool based on literature review and other patient satisfaction frameworks	The global KMO Bartlett test of sphericity was significant (p<0.0001). The Cronbach’s alpha for the scale reliability was 0.917	EFA identified 5 factors. No CFA done
Sekandi et al., 2023	Uganda	To assess the general level of patient satisfaction and identify the factors associated with the level of satisfaction among patients receiving antiretroviral therapy in Uganda	Cross- sectional study, interviewer- administered	475 patients with HIV/AIDS in Kampala, Uganda	Patient satisfaction	A modified version of the validated CAHPS instrument	KMO and Bartlett’s tests, EFA with scree plot for factor identification	No CFA. Modification of instrument done by the primary researcher, two clinicians and three research assistants. No involvement of patients or cognitive interviewing
Sendekie et al., 2023	Ethiopia	This study examined treatment satisfaction and determinant factors in patients with diabetes	Multicentre cross-sectional. Interviewer-administered	402 patients with diabetes	Patient satisfaction	DTSQ	Cronbach alpha value of 0.81	Disease specific instrument previously validated in Ethiopia. No EFA or CFA done in current study

CAHPS: Consumer Assessment of Healthcare Providers and Systems; CARE: Consultation and Relational Empathy; CFA: Confirmatory Factors Analysis; CFI: comparative fit index; cRCT: cluster randomised trial; DTSQ: Diabetes Treatment Satisfaction Questionnaire; EFA: exploratory factor analysis; GFI: goodness of fit index; GHAA: Group Health Association of America; JSE: Jefferson Scale for Empathy; JSE-S: Jefferson Scale for Empathy–student’s version; KMO: Kaiser–Meyer–Olkin; MHSSS: Mental Health Service Satisfaction Scale; MISS: Medical Interview Satisfaction Scale; MPOC: Measure of Processes of Care; NPS: net promoter score; OPD: outpatient department; PAC: post-abortion care; PACQ: Patient Assessment of Care Questionnaire; PAM: Patient Activation Measure; PCA: principal component analysis; PCAC: person-centred abortion care; PCAT: primary care assessment tool; PCAT-Mw: Malawian version of the PCAT; PCC: person-centred care; PCMC: person-centred maternity care; PEQ: Patient Experience Questionnaire; PHC: primary health care; PLWHIV: people living with HIV; PSPS: patient satisfaction with pharmaceutical services; PSS: Patient Satisfaction Survey; RMC: respectable maternity care; RMSEA: root mean square error of approximation; SAATSA: South African Addiction Treatment Services Assessment; SAPS: short assessment of patient satisfaction; SAQ: Safety Attitudes Questionnaire; SCCCS: Schwartz Centre Compassionate Care Scale; SDM-Q-9: 9-item Shared Decision-Making Questionnaire; TPS: Trust in Physician Scale; ZA-PCAT, south african primary care assessment tool.

Results

Characteristics of the included studies

Initially, 302 articles were retrieved. After the removal of duplicates, and title and abstract screening, 255 articles were excluded due to various reasons including, but not limited to, being conducted outside SSA, exploring a different concept or article type, such as review articles. A full-text review was conducted on the remaining 47 articles, of which 11 were excluded when assessed using the inclusion criteria (Figure 1). Therefore, 36 remaining articles were reviewed to understand how PCC is being measured in SSA, including the measurement tools used and the reported psychometric properties of the tools used, so as to explore their performance in the SSA setting.

Figure 1.

Prisma flow diagram showing the study selection process. SSA: sub-Saharan Africa.

Of the 36 studies, 11 were conducted in South Africa, eight in Nigeria, seven in Ethiopia, three in Tanzania, two each in Ghana and Uganda and one each from Kenya, Zambia and Malawi. Twenty-two studies (60%) were conducted in hospital settings (secondary and tertiary care), one study was conducted in a medical school and the remaining studies were conducted in primary care settings (Table 1).

An update of the search conducted on 30 December 2024 returned 34 new articles. However, none of the articles was included in the review for the following reasons: one was a study protocol, two were conducted outside of SSA (in Sweden), 28 looked at different concepts of PCC other than measurement and three were duplicates.

PCC measurement approach in SSA healthcare settings

In 35 studies, data for PCC measurement were collected using self-administered³⁶ or interviewer-administered questionnaires³⁷; or both.³⁸ Administration was to either patients within the healthcare setting or as exit interviews at the end of a healthcare visit.^39,40 Other studies used a mixed-methods approach for measuring PCC.^41,42 No studies used a focus group discussion, except in instances where a new tool was being developed.⁴² In one recent study, an online survey was used.⁴³

Audiotape recording was another approach used to measure PCC in some centres in SSA.^44,45 When audiotape recording was used, the actual assessment of PCC was conducted by external observers who rated the patient-centredness of the recorded patient–provider interaction. For example, Christoffels and Mash used the Stellenbosch University observation tool to assess the patient-centredness of the consultation skills of both doctors and nurses in primary care. Three external assessors rated the different audiotaped consultations using the observation tool that was locally developed based on the Calgary–Cambridge guide to the medical interview. The inter- and intra-rater reliability were reported to be 0.84 and 0.99, respectively.⁴⁵ No study used video recording.

In three of the included studies, PCC was measured as part of a larger study.^43,46,47 For example, Abboah-Offei and colleagues used both Patient Experience Questionnaire (PEQ) and Consultation and Relational Empathy (CARE) questionnaires when testing the efficacy and feasibility of an intervention to improve patient-centredness for people living with HIV.⁴⁷

Measurement instruments used (and their psychometric properties)

Only nine out of 31 instruments used in SSA measured PCC comprehensively. The remaining instruments either measured satisfaction or other components/dimensions of PCC, as shown in Table 2. We classified the instruments as holistic measures, domain-specific measures and satisfaction measures for presentation purposes. In addition, some of these instruments were specialty or disease specific. Below, we describe three classifications of PCC measurement instruments used in SSA.

Table 2.

Summary of instruments used to measure patient-centred care in SSA

Measurement tool	Origin	Country of development	Domain(s) measured	Target population	Validation in SSA context; country	Performance
SAPS	Hawthorne et al., 2014	Australia	Satisfaction	Outpatients attending incontinence clinics	Ratanjee-Vanmali et al., 2020; South Africa	Cronbach alpha was 0.77. No other psychometric tests are reported. Patients in an audiology clinic; 31 of 46 patients responded to the online questionnaire. Lack of internet access is a big barrier
PCMC scale	Afulani et al., 2017	Kenya	Holistic	Obstetrics	Ogbuabor et al., 2021; Nigeria	Both EFA and CFA were reported. KMO (0.95), alpha 0.94. Had acceptable goodness of fit indices (RMSR=0.032; CFI=0.908)
PEQ	Steine et al., 2001	Norway	Holistic	Primary care	Abboah-Offei et al., 2020, Nigeria	No evaluation of psychometric properties reported
CARE measure	Mercer et al., 2004	Scotland	Empathy	Primary care	Abboah-Offei et al., 2020, Nigeria	No evaluation of psychometric properties reported
Adult Primary Care Questionnaire	Hojat et al., 2011	USA	Satisfaction	Primary care	Mukamba et al., 2020; Zambia	9-item scale. EFA for construct validity (only one factor had an eigenvalue ≥1). Cronbach alpha=0.93 Criterion validity was acceptable
MISS	Wolf et al., 1978	USA	Satisfaction	Primary care	Eksteen et al., 2019, South Africa	21-item scale. Overall reliability 0.89. Subscales such as communication had very low reliabilities (0.09–0.54). No factor analysis was done
PAM-13	Hibbard et al., 2004; and Hibbard et al., 2005	USA	Activation/engagement	Patients with chronic disease	Siddharthan et al., 2016; Uganda	13-item scale. No performance evaluation conducted
MHSSC	Mayston et al., 2017, in Ethiopia	Ethiopia	Satisfaction	Psychiatric patients	Kassaw et al., 2022; Ethiopia	24-item tool. Kassaw et al., 2022, did not report on psychometric performance Mayston et al., 2017, performed EFA, reported overall alpha of 0.89, with inter-item correlation 0.27–0.31
TPS	Anderson and Dedrick, 1990	USA	Trust	Patients with diabetes	Aloba et al., 2014; Nigeria	11-item scale. Weak psychometric performance reported. Conducted EFA using PCA with varimax rotation. Items loaded on 2 factors (loadings were >0.40). Alpha 0.68
PSPS questionnaire	Njilele et al., 2012 in Nigeria	Nigeria	Satisfaction	Patients with HIV/AIDS	Njilele et al., 2012 (Nigeria); Okoye et al., 2014 (Nigeria)	16-item scale. Alpha 0.85 (range of 0.66–0.81 for the subscales). PCA and construct validity (EFA revealed a 4-factor structure). Spearman rho correlation for convergent and discriminant validity were acceptable
SERVQUAL (see Khamis et al., 2014)87	Parasuraman et al., 1988	USA	Satisfaction	Consumers’ perception of service quality	Khamis et al., 201487 (Tanzania); Kassa et al., 2021	26-item scale. KMO was 0.87, EFA using PCA, overall alpha 0.87, range 0.78–0.83 (Khamis et al.),87 0.87, range 0.75–0.77 for subscales (Kassa et al.)
Patient Satisfaction Survey	Afolabi et al., 2012, Nigeria	Nigeria	Satisfaction	Outpatients, pharmaceutical services	Afolabi et al., 2012; Nigeria	25-item scale. KMO=0.80, EFA using PCA revealed a 6-factor structure. Inter-item correlation ranged from 0.43 to 0.75, alpha 0.71 and Guttman split half coefficient was 0.79
Measures of Processes of Care	King et al., 2004	Canada	Holistic	Carers of children with disability	Saloojee et al., 2009; South Africa	20-item scale found to have weak psychometric properties. Alpha 0.24–0.66; ICC 0.51–0.61
DTSQ	Bradley et al., 1994 (Handbook of Psychology and Diabetes)	UK	Satisfaction	Patients with diabetes	Kolawole et al., 2004 (Nigeria); and Sendekie et al., 2023	The 8-item scale had acceptable performance. Cronbach alpha of 0.81 Alpha 0.74 (Kolawole), inter-item correlation 0.22–0.79; item-total correlation 0.39–0.78
PACQ	Myers et al., 1990	UK	Satisfaction	Psychiatric patients	Olusina et al., 2002; Nigeria	16-item scale with 6 domains. Negative items had to be avoided during the back translation to Yoruba language. Alpha 0.79; face and content validity assessed
Patient-Centredness in the Consultation	Henbest and Stewart, 1989, Canada	Canada	Holistic	Primary care	Henbest and Fehrsen, 1992; South Africa	Depends on external observers rating an audiotaped consultation. Strong inter-rater reliability (0.87–0.95). Efficient: the 2-min score correlated highly with the score for the entire consultation
PCAT	Shi et al., 2001	USA	Holistic	Primary care	Dullie et al., 2018; Malawi	The 29-item scale (PCAT-Mw) developed by adaptation of the South African version ZA-PCAT (Bresick et al., 2015) performed acceptably well. Good face and content validity, internal consistency (alpha 0.66–0.91), ICC=0.90, satisfactory goodness of fit (GFI=0.90, CFI=0.91, RMSEA=0.05, PCLOSE=0.65)
SCCCS	Lown et al., 2015	USA	Compassion	Psychiatric patients	Abate et al., 2023; Ethiopia	Psychometric performance of the 12-item scale not discussed
JSE-S	Hojat et al., 2001	USA	Empathy	Medical students	Archer et al., 2019; South Africa	The 20-item scale had a reliability of 0.81. Good performance overall. Three negatively phrased items performed poorly with absence of discriminatory ability
SAATSA	Myers et al., 2015	South Africa	Holistic	Drug addicts	Myers et al., 2015; South Africa	Psychometrically robust tool; 31-item scale with 6 subscales. Only 2 subscales explored patients’ perceptions of access and quality of care. CFA with goodness of fit indices: CFI=0.98, RMSEA=0.06 (0.04–0.09). Alpha 0.72–0.89
PCAC scale	Sudhinaraset et al., 2020.	Kenya	Holistic	Obstetrics (PAC)	Sudhinaraset et al., 2020; South Africa	The 24-item scale (with two subscales) had high content, construct and criterion validity. Scale stratified into surgical abortion (alpha 0.72–0.82) and medical abortion (alpha 0.65–0.82)
American Pain Society Patient Outcome questionnaire	Gordon et al., 2010	USA	Satisfaction (perspectives on postoperative pain management)	Postoperative surgical patients	Woldehaimanot et al., 2014; Ethiopia	13-item scale. EFA with PCA that extracted 3 factors, overall Cronbach alpha was 0.78. No other psychometric analyses reported
CAHPS	Crofton et al., 1999	USA	Holistic	Patients at all levels	Sekandi et al., 2023; Uganda	Modified the CAHPS into an 18-item scale. Good performance although no patient involvement or cognitive interviewing during instrument modification. KMO=0.94, alpha=0.94
Others—tools locally developed in SSA but without a specified name
Author, year		Country of development	Domain/dimension assessed	Target population	Validation	Performance
Baynes et al., 2019		Tanzania	Satisfaction	Obstetrics (PAC)	Baynes et al., 2019	12 questions for ranking the level of satisfaction with dimensions of PAC quality, including waiting time, privacy, cleanliness and treatment from staff and the PAC providers. Each question scored 1–4 (1, lowest; 4, highest). Alpha 0.81. Tool takes about 30–45 min to complete
Clarke-Deelder et al., 2022		Tanzania	Holistic	Caregivers of sick children	Clarke-Deelder et al., 2022	Questions assessed user experience across three domains: prompt care, respect and communication. No adequately performing factor structure after EFA so a single 8-item additive measure was proposed (without any subscales). Reliability (alpha) was 0.7
Kagura et al., 2023		South Africa	Satisfaction	Patients with chronic disease in primary care	Kagura et al., 2023	The 22-item scale with 5 subscales (improving values and attitudes, cleanliness in the clinic, safe and effective care, infection control, availability of medicines) performed well. Items derived from existing literature. KMO was 0.90 and EFA showed high factor loadings. Alpha 0.917; no CFA was conducted
Sheferaw et al., 2016		Ethiopia	RMC	Obstetrics	Sheferaw et al., 2016	15-item scale with 4 subscales (friendly care, abuse-free care, timely care, discrimination free care). EFA showed high factor loadings with the subscales (range 0.76–0.82). Alpha 0.85
Agu et al., 2014		Nigeria	Satisfaction	Patients with HIV/AIDS	Agu et al., 2014	No evaluation of psychometric properties reported
Westaway et al., 2002		South Africa	Satisfaction	Black patients with diabetes	Westaway et al., 2002	25-item scale with 3 factors following EFA (support + communication, service logistics and technical expertise). Reliability was 0.98; Validity using multitrait tests
Westaway et al., 1998		South Africa	Satisfaction	Family planning services	Westaway et al., 1998	17-item scale with 2 factors (interpersonal and organisational dimensions) accounting for 51% variance following EFA. Alpha was 0.76
Ulys et al., 1997		South Africa	Satisfaction	Psychiatric patients	Ulys et al., 1997	Multiple questions derived from literature to assess quality using consumer expectations as well as provider ratings. Not much psychometric assessment. Inter-rater reliability was 0.94

CAHPS: Consumer Assessment of Healthcare Providers and Systems; CARE: Consultation and Relational Empathy; CFA: xxxxxx; CFI: comparative fit index; DTSQ: Diabetes Treatment Satisfaction Questionnaire; EFA: exploratory factor analysis; GFI: xxxx; ICC: intra-class correlation; JSE-S: Jefferson Scale for Empathy–student’s version; KMO: Kaiser–Meyer–Olkin; MHSCC: Mental Health Service Satisfaction Scale; MISS: Medical Interview Satisfaction Scale; PAC: post-abortion care; PACQ: Patient Assessment of Care Questionnaire; PAM: Patient Activation Measure; PCA: principal component analysis; PCAC: person-centred abortion care; PCAT: primary care assessment tool; PCAT-Mw: Malawian version of the PCAT; PCLOSE: xxxx; PCMC: person-centred maternity care; PEQ: Patient Experience Questionnaire; PSPS: patient satisfaction with pharmaceutical services; RMC: respectable maternity care; RMSEA: root mean square error of approximation; RMSR: xxxx; SAATSA: South African Addiction Treatment Services Assessment; SAPS: short assessment of patient satisfaction; SCCCS: Schwartz Centre Compassionate Care Scale; SSA: sub-Saharan Africa; TPS: Trust in Physician Scale; ZA-PCAT, xxxx.

Holistic measures

Holistic measures refer to instruments that comprehensively measure PCC as a concept. Of the nine holistic measures identified in this review, two are focused on PCC in obstetrics and gynaecology (person-centred maternity care [PCMC] and person-centred abortion care [PCAC]),^25,48 four on primary care,^44,47,49,50 two on caregivers of sick children^40,51 and one on psychiatric patients.⁵²

Four instruments—PCMC, PCAC, South African Addiction Treatment Services Assessment [SAATSA] and that by Clark-Deelder and colleagues—were locally developed,^25,40,48,52 whereas the others were developed in Norway (PEQ),⁵³ the UK (MPOC),⁵¹ the USA (primary care assessment tool [PCAT] and Consumer Assessment of Healthcare Providers and Systems [CAHPS])^50,54 and Canada (PC in Consultation).⁴⁴ The most common domains assessed were communication (PEQ, PC in Consultation)^44,47, patient-provider relationship, continuity and coordination of care (Malawian version of the PCAT [PCAT-Mw]),⁴⁹ autonomy, dignity and respectful care (PCMC, PCAC).^25,48 These were assessed using a total number of items ranging from eight to 31, which were divided into subscales. The only exception was the measure by Clark-Deelder et al.,⁴⁰ which includes an eight-item scale without any subscales.

Only three instruments (PCAT, SAATSA and PCMC) included a comprehensive assessment of goodness of fit indices during their validation.^49,52,55 These studies reported on the root mean square error of approximation, comparative fit index and the Tucker–Lewis Index. Although the studies reported face and content validity as well as assessment of construct validity through exploratory factor analysis, there was no formal assessment of convergent and discriminant validity of the instruments in the population in SSA. Locally developed instruments, such as, for example, SAATSA, PCMC and PCAC,^25,48,52 displayed a better psychometric performance compared with instruments developed in the Global North and adapted for use in SSA (MPOC, CAHPS and PC in consultation).^27,44,56

Domain-specific measures

These instruments measured specific subcomponents of PCC and included the Trust in Physician Scale (TPS),⁵⁷ Jefferson Scale for Empathy (JSE),⁵⁸ the Schwartz Centre Compassionate Care Scale,⁵⁹ Patient Activation Measure (PAM),⁴⁶ the American Pain Society Patient Outcome Questionnaire (APS POQ)⁶⁰ and the CARE scale.⁴⁷

The original scales were mostly developed in the USA (TPS, JSE, PAM and the APS POQ)^61–64 and in the UK (CARE),⁶⁵ for specific populations, notably for patients with diabetes (TPS)⁶¹ and for patients with chronic diseases (PAM).⁶³ However, instruments have been used in patient populations within SSA that were different to those in which they had been developed. For example, the TPS first developed by Anderson and Dedrick, in 1990, among patients with diabetes in the USA as a measure to determine patients’ trust in their primary physician,⁶¹ was used to assess trust among psychiatric patients in Nigeria. The original scale consisted of 11 items scored on a five-point Likert scale and contained questions that explored patients’ extent of trust in the physician’s counsel, judgement and medical treatment preference. In their analyses, the authors reported the TPS to have weak psychometric properties with a Cronbach’s alpha of 0.68 among Nigerian psychiatric patients.⁵⁷

Furthermore, several studies did not report on the psychometric properties of instruments.^46,47,59,66 However, Siddharthan et al. report that the PAM-13 scale was able to discriminate or distinguish between predefined study subgroups such as private and public patients.⁴⁶

Measures of satisfaction

Seventeen out of 31 instruments (55%) measured satisfaction as a way of assessing the patient-centredness of healthcare from the perspective of the patient. These measures were either general (e.g. SERVQUAL, Patient Satisfaction Survey [PSS], adult primary care questionnaire, short assessment of patient satisfaction [SAPS])^43,67–69 or were disease or subspecialty specific, measuring service dimensions of psychiatry,^70,71 obstetrics and gynaecology,^37,42 diabetes,^72,73 pharmaceutical services (e.g. patient satisfaction with pharmaceutical services [PSPS])^74,75 or HIV.⁶⁶

Only two instruments were developed based on a theoretical understanding of satisfaction, thereby facilitating measurement of specific dimensions of satisfaction.^43,76–78 For example, the 26-item medical interview satisfaction scale validated by Eksteen and Mash in South Africa is based on cognitive, affective and behavioural aspects of satisfaction.⁷⁷ Similarly, the SAPS is based on the theory that patient satisfaction has seven dimensions, consisting of access to health services, provision of health information, patient–provider relationship, shared treatment decision-making, technical quality of care received, the extent to which treatment meets patients’ expectations and general satisfaction.^43,78 These dimensions can be considered components of patient–provider interactions, and therefore relevant to PCC.

Development and validation

Eight instruments were developed and validated in SSA using various methods.^{37,41,42,66,68,72,74,79} The commonest methodological approach was development of questions from a prior literature review, followed by face and content validity assessment through an external expert review of questions.^41,68,74,79 Only the Mental Health Service Satisfaction Scale was developed with themes from a qualitative study.⁴¹ Afolabi et al. conducted a pilot survey with 30 participants as part of the development process of the PSS scale. This allowed them to refine the scale and reduce it from 35 to 25 items.⁶⁸

Psychometric evaluation was either incomplete in several studies^43,77,80 or not performed in others,⁵⁹ whereas some studies evaluated and reported good psychometric performance.^49,52,55,74 For example, the 16-item PSPS had an overall Cronbach alpha of 0.85 with good convergent and discriminant validity assessed using the Spearman rho correlation method,⁷⁴ whereas Westaway et al. assessed validity using the multitrait method.⁷² However, some instruments did not follow this approach, but rather were developed as study-specific questionnaires.^37,42,66,72 The number of PCC dimensions assessed ranged from three to six, and included factors such as technical quality, logistics, interpersonal relationship, timeliness of services, availability of medicines, access to health services, shared treatment decision-making, patient–provider relationship and communication.

Discussion

The current review aimed to explore how PCC is being measured in healthcare settings in SSA and to describe the psychometric performance (where tested and reported) of instruments used to measure PCC in these settings. Thirty-one instruments were identified (Table 2). Nine instruments measured PCC holistically, albeit in specific patient populations. Overall, the psychometric performance of the identified instruments was inconsistently reported. The most common inconsistency was in measurement variance and structural validity of instruments. The most common approach in the measurement of PCC was using interviewer-administered questionnaires.

The current review’s findings are similar to those from reviews conducted in the Global North,^4,29–31 which reported multiple instruments being used in the measurement of PCC, and inconsistent and incomplete reporting of psychometric properties. The importance of this review lies in its focus on the availability of instruments developed and/or used for measuring PCC in the Global South. Similar to the 2014 review conducted by the UK Health Foundation, where only five of 160 tools measured all components of PCC,⁴ this review, like Kumah’s on satisfaction with care,¹⁵ found that most instruments used in SSA do not directly measure PCC.

Such heterogeneity may partly be due to differences in the framework and definition of PCC used by the authors.⁸¹ Moreover, this review identified a parallel development of new instruments for measuring the same service. For example, Afolabi et al. and Njilele et al. both separately developed instruments to measure patient satisfaction with pharmaceutical services in Nigeria in the same year.^68,74 This overlap may have been due to the lack of an extensive literature search and expert review, which forms an important step in instrument development.⁸² This reinforces the importance of this review for those contemplating undertaking PCC measurement studies in SSA.

Whereas measures of patients’ satisfaction—the most common approach—are used to inform healthcare quality improvements, studies limited to this aspect do not comprehensively evaluate the care process and therefore do not specify which areas in the care process need to be improved.¹² This, in turn, could potentially misinform clinicians and policymakers and mislead quality improvement interventions. For example, whereas the PCMC scale would provide details such as empathy, dignity, autonomy and confidentiality in the process of care, the PSS would only give an overall outcome (satisfaction) score without any actionable areas of improvement. PCC studies in SSA are relatively rare compared with the Global North,^15,30,83 probably due to a lack of funding, an absence of patient advocates to push for such studies and/or a lack of research capacity or a community of researchers with the skills to conduct such studies. A recent review reported that most of the instruments used in the measurement of patient experience in SSA used hospital-level rather than individual-level data.¹⁵

On a positive note, despite psychometric properties being inconsistently reported, instruments used in Malawi and Nigeria, such as the PCAT-Mw,⁴⁹ PSPS⁷⁴ and the PCMC,⁵⁵ had relatively extensive assessment and acceptable psychometric properties. However, the heterogeneous and inconsistent performance of PCC measurement instruments in SSA suggests cultural differences between SSA and the Global North, where most of these instruments were developed,²⁷ or low literacy levels in SSA.⁸⁴ For example, while evaluating the MPOC-20, Saloojee et al. noted that, in cross-cultural and resource-poor settings, the process and the words used in asking questions had to be changed in order to adapt the MPOC tool.²⁷ This is in line with current recommendations regarding cross-cultural adaptation of tools that involve, among others, backward and forward translation.⁸⁵ This practice, however, was not reported in the studies included in this review.

The combination of low literacy levels, low healthcare service quality expectations and ‘a desire to please’ in responding to questions in SSA settings may explain the paradoxically high rating of satisfaction and patient experience, despite low scores in the process of receiving care, including domains such as choice, autonomy and attention, as well as clinical outcomes.^15,86 Such cultural factors, if not considered in the development and local validation of such tools, may undermine the goal of PCC measurement in SSA, leading to complacency due to patients giving high scores on the domains of care received, whereas patient engagement (with the research process and purpose), attention to patient empowerment and the development and testing of comprehensive tools within SSA settings could put patients into a better position to critique and drive improvements in their health system.

The local development and testing of bespoke tools are critically important. Some PCC measurement instruments included in this review were used in populations for which they were not developed. For example, the TPS for patients with diabetes in the USA was validated among psychiatric outpatients in Nigeria by Aloba et al.⁵⁷ In their analyses, they report the TPS to have weak psychometric properties with a Cronbach’s alpha of 0.68. This highlights the importance of the development and/or validation of tools in the appropriate population.

The predominant data collection method used was an interviewer-administered questionnaire. One study used an online survey in the measurement of PCC.⁴³ This is also most likely due to low literacy levels and limited access to the internet in most communities in SSA. This may partly explain why the study employing an online survey included only 31 participants. Therefore, methods involving interviewer-administered questionnaires that do not place a significant cognitive load on participants are most suitable for measuring PCC in SSA.

A limitation of this review is that although an attempt was made to include all possible studies conducted at all healthcare levels in SSA, some relevant articles that discuss the measurement of PCC may have been omitted due to publication bias. Relevant articles in French or Portuguese may also have been omitted. We did not assess the methodological quality of the included studies. Therefore, the evidence was not appraised for its relevance, reliability and validity. However, the use of a scoping review has enabled a general overview of the current state of PCC measurement in SSA. Although Table 2 brings out some of the significant elements in the COSMIN criteria, not all domains were captured.

In conclusion, heterogeneity and a lack of comprehensive approaches to measuring PCC in SSA, along with inconsistent and incomplete reporting of the psychometric performance of these instruments, mean that national policymakers and managers lack the critical evidence needed for promoting PCC. We recommend the following measures to address this: use of instrument reporting guidelines during validation and/or cultural adaptation of tools,⁸⁵ and research funding and capacity building, including South-to-South initiatives, to support the conduct and reporting of instrument development studies in SSA.^33,82 A significant challenge to overcome is developing ways in which PCC is measured and reported by patients without bias. In addition, more general and inclusive holistic approaches to measuring PCC in SSA are needed, especially in subpopulations in which such studies have not yet been conducted or explored. For example, currently no suitable instrument exists for the measurement of PCC among surgical patients in SSA.

Appendix 1 Search strategy for the scoping review.

All searches are from 1990 to September 2024, except for the Global Health database, which was from 1973–2024. Web of Science (up to 22 September 2024).

All searches were updated on 30 December 2024.

#1	ALL=("patient-centred care" OR "person-centred care" OR "individualised care" OR "family-centred care" OR "patient experience" OR "patient satisfaction")	66 735
#2	ALL=("measurability" OR "measurable" OR "measurably" OR "measures" OR "measureable" OR "measured" OR "measurement" OR "measurement s" OR "measurements" OR "measurer" OR "measurers" OR "measuring" OR "measurings" OR "measurment" OR "measurments" OR "weights and measures" OR "measure" OR "measures" OR "tools" OR "assess" OR "assessement" OR "assesses" OR "assessing" OR "assessment" OR "evaluation")	14 196 226
#3	ALL=("valid"[AOR "validate"[All Fields]15 OR "validated"[All Fields] OR "validates"[All Fie"validation"OR "validity"OR "reliability"OR "feasibility"OR "psychometric propert*")	1 825 204
#4	ALL=("sub saharan africa"OR"africa south of the sahara")	55 020
#5	#1 AND #2 AND #3 AND #4	5

PubMed/Medline (Up to 21 September 2024).

((((("patient-centred care") OR ("person-centred care")) OR ("individualised care")) OR ("family-centred care")) OR ("patient experience")) OR ("patient satisfaction")	125 004
((((measurement) OR (tools)) OR (measur*)) OR (assessment)) OR (evaluation)	10 061 930
((((validity) OR (reliability)) OR (feasibility)) OR ("psychometric propert*")) OR (validation)	1 772 222
(("sub saharan africa") OR (africa)) OR ("africa south of the sahara")	441 213
((((((((("patient-centred care") OR ("person-centred care")) OR ("individualised care")) OR ("family-centred care")) OR ("patient experience")) OR ("patient satisfaction")) AND (((((measurement) OR (tools)) OR (measur*)) OR (assessment)) OR (evaluation))) AND (((((validity) OR (reliability)) OR (feasibility)) OR ("psychometric propert*")) OR (validation))) AND ((("sub saharan africa") OR (africa)) OR ("africa south of the sahara")))	197

Embase Session Results (Up to 21 September 2024).

No. Query	Results
#5 #1 AND #2 AND #3 AND #4	83
#4 'sub saharan africa' OR africa OR 'africa south of the sahara'	379 081
#3 validity OR reliability OR feasibility OR 'psychometric propert*' OR validation	1 290 682
#2 measurement OR tools OR measur* OR assessment OR evaluation	9 826 590
#1 'patient-centred care' OR 'person-centred care' OR 'individualised care' OR 'family-centred care' OR 'patient experience'/exp OR 'patient experience' OR 'patient satisfaction'/exp OR 'patient satisfaction'	194 734

Global Health <1973 to 2024>

1	patient-centred care or person-centred care or individualised care or family-centred care or patient experience or patient satisfaction).mp. [mp=abstract, title, original title, heading words, cabicodes words]	3684
2	(measurement or tools or measur* or assessment or evaluation).mp. [mp=abstract, title, original title, heading words, cabicodes words]	1 133 931
3	(validity or reliability or feasibility or psychometric propert* or validation).mp. [mp=abstract, title, original title, heading words, cabicodes words]	101 892
4	("sub saharan africa" or africa or "africa south of the sahara").mp. [mp=abstract, title, original title, heading words, cabicodes words]	285 801
5	1 and 2 and 3 and 4	17

Author contributions

Paul K. Okeny (Conceptualization, Data curation, Formal Analysis, Writing—original draft), Chiara Pittalis (Data curation, Formal Analysis, Writing—review & editing), Jakub Gajewski (Data curation, Formal Analysis, Writing—review & editing), and Ruari Brugha (Formal Analysis, Writing—review & editing)

Funding

This work was supported by the Strategic Academic Recruitment (StAR) International PhD programme of the Royal College of Surgeons in Ireland, Dublin, Republic of Ireland.

Competing interests

The authors declare no competing interests.

Ethical approval

Not applicable.

Data availability

All data are available on request.