Table 4.

Qualities and psychometric properties of the selected generic measures extracted from the reviews

Scale	Domains addressed	Completion time/Number of items	Response format	Accessibility	Validity	Reliability	Responsiveness	Floor/ Ceiling effects	Conclusions/recommendations of the reviews
COOP/WONCA [21]	Physical fitness, feelings, daily activities, social activities, pain, change in health, overall health, social support and quality of life	<5 mins [58] 9 items (more recently reduced to 6)	Five options with pictorial depictions accompanying the text	Available in 20 languages [18] Acceptable and feasible [13]	HIV positive women had poorer scores than HIV negative women on six out of nine health dimensions (construct validity) [13]	not stated	not stated	not stated	not stated
EQ-5D [22, 23]	Mobility; self-care, usual activities, pain/discomfort, anxiety/depression, self-reported health	1 min [4] 6 items	5 dimensions of quality of life are rated on either a 3 point scale (no problems/ some or moderate problems/ extreme problems – EQ-5D-3 L) or 5 point scale (no problems/ slight problems/ moderate problems/ severe problems/ extreme problems – EQ-5D-5 L). In addition, a visual analogue scale (0–100) is used to rate overall health.	Approximately 1 min to complete [4]. Available in multiple languages [12, 18]. Can be administered electronically or over phone [12]. General population preference weights have been derived for many countries [4].	Correlates with MOS-HIV subscales and discriminates between participants stratified by HIV/AIDS severity based on CD4 count/viral load (construct validity) [4, 12]. Lower EQ-5D scores among people with HIV not receiving ART than general population (construct validity) [13].	not stated	Responsive to initiation of ART, the development of opportunistic infections and adverse events [4, 10, 12, 13], with small-to-medium effect sizes in each of its five dimensions [10].	Ceiling effects in general population samples [4, 12, 20].	Clayson et al. recommend using the EQ-5D alongside a disease-targeted measure, however because of ceiling effects in general population samples they would not recommend the EQ-5D for studies including individuals with early, asymptomatic HIV infection. [4] Wu et al. recommend use alongside the MOS-HIV to obtain HIV-specific HRQL and utility measures [10]. Performance equivalent to the MOS-HIV in clinical trials [20]. Can generate indirect health utility values for use in economic models [10, 11].
FLZM Questions on life satisfaction [24]	Satisfaction with life in general: friends’ free time, general health, financial security, work, life conditions, family life and relationships. Satisfaction with health: physical condition, ability to rest, energy, mobility, freedom from anxiety, freedom from pain, independence	“A few minutes” [24] 16 items	5 point scales rating the importance of and satisfaction with each aspect of quality of life.	not stated	not stated	not stated	not stated	not stated	not stated
Health Utilities Index (HUI) HUI2; HU13 [25]	Vision, ambulation, dexterity, emotion, cognition, hearing, speech and pain	5–10 mins [59] 15 items	4–6 response options for each question	Available in multiple languages [12]. Can be administered electronically [12].	HUI2 and 3 have been associated with disease severity/AIDS related events and plasma viral load (construct validity). [4] Correlates well with most MOS-HIV subscales (convergent validity) [4, 12, 20].	not stated	Responsive to change in HIV disease states [4, 12, 20], however the MOS-HIV and the EQ-5D VAS had better discriminatory capacity [12].	not stated	Despite less evidence for the HUI than EQ-5D and SF-36, emerging data were positive. [4] Potentially useful adjuvant to an HIV-specific measure in a trial. [4]
McGill Quality of life questionnaire (M-QOL) [26]	Physical, Psychological, Existential, Support.	16 items	Two response options for each item (e.g. no problem vs tremendous problem)	not stated	Content/face validity: The existential dimension is particularly relevant to people with advanced disease (CD4 < 100) [20] Only scores for physical symptoms distinguished between people with HIV with low and high CD4 count (construct validity). [20]	Factor analysis indicated four reliable subscales plus a single item about physical wellbeing (internal consistency). [20]	not stated	not stated	not stated
SF-12 [27]	Physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional, mental health	2–3 mins [60] 12 items	2–6 response options per item	not stated	not stated	No internal consistency data reported [17].	Mixed results in terms of responsiveness to change in treatment [12].	Likely to have similar floor and ceiling effects to other MOS measures [17].	Clayson et al. recommend the use of the SF-12 where the length of the SF-36 is a problem. [4]
SF-20 [28]	not stated	3–5 mins [61] 20 items	3–6 response options per item	not stated	No construct validity data [17].	Adequate cronbach’s alphas (internal consistency) [17].	not stated	Floor and ceiling effects noted in some dimensions [17].	not stated
SF-36 [29–31]	Physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, mental health, reported health transition	7–10 mins [4] 36	2–6 response options per item	Has been translated into several different languages [18, 20]. Dominates generic HR-QOL measurement with normative scores for US, UK and many other countries [4, 20]. Takes 7–10 min to complete [4]. Can be administered electronically [18]	Correlates with disease severity, CD4 counts and other measures of QOL.(construct validity) [12] PLWL reported lower QOL on all dimensions, compared to healthy controls [12, 13, 20], with the biggest decline between Stages 1 and 2 of the disease (construct validity) [13, 20]. Scale scores were associated with treatment duration, less co-morbidity, and better social support improved physical functioning (construct validity) [13].	Cronbach’s alpha within acceptable range (internal consistency) [12, 17].	Responsive to the initiation of ART and change in CD4 count, viral load and the number of symptoms. [4] Improvement in all HRQOL domains along with clinical indicators after starting ART. May not be sensitive to change of ART medication in people with HIV who are stable on ART [12].	Problems with floor and/or ceiling effects for some subscales [17, 20].	More evidence for the SF-36 in people with HIV than other recommended generic measures (EQ-5D or HUI) and the SF-12 is a viable alternative if the length is a problem. [4] Coluatti et al. recommend the SF-36 as the most appropriate generic measure for assessing HRQL in people with HIV [18]. Use alongside a disease-targeted measure (other than the MOS-HIV which shares items) is recommended. [4] Unclear whether there is an advantage to using the MOS-HIV over the SF-36. [4] This and other MOS measures were developed in US – although translated into other languages people from these countries had no input into development and these versions may have limited semantic equivalence [20]. Can be used in cost-utility analyses by deriving utility weights from the SF-36 [11].
WHOQOL-BREF [32, 33]	Physical health, psychological health, social relationships and environment.	<5 mins [20] 26 items	5-point scales	Available in 40 languages. Takes <5 min to complete [20]. Developed in 15 centres worldwide to increase cross-cultural validity [12].	Correlates well with disease severity, patients who had lower CD4 counts had lower HRQOL (construct validity) [12].	Cronbach’s alpha coefficients in the acceptable range (internal consistency) [12].	not stated	not stated	Developed from the WHOQOL-100 measure, which was developed within an international collaboration of 15 countries using a spoke-wheel methodology to ensure conceptual and semantic equivalence [20].