Table 5.

Qualities and psychometric properties of the selected HIV-specific measures extracted from the reviews

Scale	Domains addressed	Completion time/No. items	Response format	Accessibility	Validity	Reliability	Responsiveness	Floor/ Ceiling effects	Conclusions/recommendations of the reviews
ACTG SF-21 [34]	Physical functioning, energy/fatigue, social functioning, role functioning, cognition, pain, health perception and emotional well-being.	4–5 mins [34] 21 items	3–6 response options per item plus a visual analogue scale	Available in 2 languages [17].	No validity data available [17].	No internal consistency reported [17].	not stated	Likely to have similar floor and ceiling effects to other MOS measures [17].	not stated
HIV-QL31 [35]	Sex life, pain, psychological aspects, relationships, limitations caused by HIV, symptoms, impact of treatment and care	31 items	Yes/No	Available in English and French [18].	Discriminates between groups with different severity levels (construct validity) [20].	High internal consistency [17, 20].	not stated	not stated	QL-31 is a relatively sound and useful instrument where attention has been paid to the breadth of the concept as a result of listening to the concerns of patients. However it is culture-specific in the ways it has been designed [20].
MQOL-HIV [37]	Mental health, physical health, physical functioning, social functioning, social support, cognitive functioning, financial status, partner intimacy, sexual functioning, perceived access to medical care	10 mins [4] 40 items	Likert scale (never – always)	In a study comparing the MQOL-HIV with the MOS-HIV there were more missing data/incomplete responses on the MQOL-HIV [12]	Discriminates between patients based on symptom severity, inpatient care and stage of illness [4]. MQOL-HIV scores distinguished between AIDS, symptomatic HIV, and asymptomatic cases, on 7 domains and overall QOL in one study however inadequate discrimination between disease stages was found in a Spanish study [20].	Good internal consistency (Cronbach’s alpha >0.70) for 8 of the 10 domains [4]. Poor internal consistency for physical and mental health, physical and sexual functioning [20]. Good test-retest reliability for all domains except cognitive functioning [4]. Poor test-retest reliability [20].	Somewhat responsive to change in the number of symptoms, viral load and CD4 count over a 3-month period [4]. Responsive to perceived changes in quality of life over 6 months in one study however only five dimensions were sensitive to clinical changes during ART [20]. Less sensitive than the MOS-HIV for detecting changes after starting or switching ART. MOS-HIV detected change on a greater number of subscales [12].	Floor and/or ceiling effects were reported in some dimensions [17]. Fewer problems with floor ceiling effects when compared to the MOS-20 [20].	The MQoL-HIV was not one of the reviewed measures recommended by Clayson et al. [4] A relatively sound and useful measure where attention has been paid to patient input and the breadth of the concept however the instrument is culture-specific [20].
PROQOL-HIV [38, 41]	8 scored domains: Physical health and symptoms, treatment impact, emotional distress, health concerns, body change, intimate relationships, social relationships, stigma and 4 additional items addressing religious beliefs, finance, having children and satisfaction with care.	7 mins [51] 43 items	Rated on a 5-point scale ranging from 0 = ‘never’ to 4 = “always”	not stated	not stated	not stated	not stated	not stated	No information reported in the reviews
MOS-HIV [36]	Two summary scores—the physical health score and mental health score and 10 domains: physical functioning, pain, social functioning, role functioning, emotional well-being, energy/fatigue, cognitive functioning, health distress, health transition, general health and overall quality of life	10 mins [4, 12] 35 items	2–6 response options per item	Translated into at least 14 languages, largely designed for industrialised world [20]. English version takes approx. 5–10 min [4, 12, 20] but twice as long has been reported for the use of some translations, e.g. Spanish, where more words are needed to express the concepts [20]. Scoring/ interpretation is complex [12, 20]. Less missing data than MQOL-HIV [12].	Mixed reports regarding construct validity with some suggesting poor construct validity [17] and others suggesting good construct validity [20]. Large body of evidence supporting convergent and discriminant validity [4, 12], although some studies have not found the expected relationship with CD4 count (construct validity) [12, 20].	Moderate /good internal consistency generally reported [4, 12, 19, 20] although Carabin reported good internal consistency for some but not all domains [17]. Inadequate test-retest reliability [17, 19].	Responsiveness has been established in a wide variety of contexts including adverse events, increased symptoms, opportunistic infections and AIDS-defining events, initiation of ART [4, 10, 12, 18]. Negligible effects in treatment experienced adults changing therapy [10]. Studies have found the MOS-HIV is more responsive than EQ-5D, HUI3 and MQOL-HIV [12].	Floor and ceiling effects have been reported [19, 20].	Well established reliability/validity and widely used but concerns that as one of the earliest HIV-specific measures to be developed it may not have continued relevance for PLWH. They question whether there is a true advantage to using the MOS-HIV over the SF-36. Would be unwise to administer alongside another MOS measure such as the SF-36 because of shared items [4]. May have limited value in cross-cultural research because although the scale has been translated into many languages, it may not have sematic and conceptual equivalence outside the USA [20]. More information is needed about performance of the scale in women, low income and other socially disadvantaged groups [20]. Can be administered individually or together with the EQ-5D to measure changes in HRQOL [10]. There is a lack of items addressing sexual function and body image [18]. Validation data draws on a range of patient groups from asymptomatic to those with advanced HIV [4].
WHOQOL-HIV BREF [39]	Physical, psychological, level of independence, social, environmental, and spiritual QoL	31 items	5 point likert scale	not stated	not stated	not stated	not stated	not stated	not stated
Symptom Quality of Life Adherence (HIV-SQUAD) [40]	HRQOL items include physical and psychological domains. The measure also includes symptoms and a visual analogue scale for adherence	24 items	5 point likert scales, dichotomous items and a visual analogue scale	not stated	The measure discriminated between patients at different CD4 counts and with/without hepatitis co-infection (construct validity) [12].	Cronbach’s alpha was acceptable for the physical component but <0.7 for the psychological component (internal consistency) [12]	Responsive to change in HIV viral load [12].	not stated	not stated