Table 7. Main characteristics of studies evaluating the cost-effectiveness of interventions primarily aimed at supporting family caregivers.
| No. | Study | Intervention description and comparator (Setting) | Country | Type of study and economic evaluation (Time horizon) | Sample size No. PwDs/No. Caregivers (Intervention Group/Control Group) |
Perspective | PwD outcome measures | Caregiver outcome measures | Incremental Cost-Effectiveness Ratio (ICER) and other cost-effectiveness measures | Cost-effective. Assess. | Cost-effectiveness rationale | Quality of the study |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1.1 | Knapp et al. 2013 [71] | Individual therapy program for informal caregivers (START) vs Usual Care alone (Community-based: mental health and neurological outpatient dementia services) |
UK | RCT CEA+CUA (8 months) |
260 / 260 (173 / 87) |
Health and social care system |
|
|
In the base-case analysis (only caregiver’s costs): £6,000 per QALY gained (caregiver); £118 per HADS point (caregiver) | High | The short-term intervention had a 99% probability of being cost-effective for carers at the NICE’s WTP threshold of £30,000 per QALY gained. Moreover, START showed a high probability of cost-effectiveness on the HADS-T (Hospital Anxiety and Depression Scale) measure even though the authors were not aware of societal WTP for gauging cost-effectiveness on the HADS scale. | High |
| 1.2 | Livingston et al. 2014 [117] | Individual therapy program for informal caregivers (START) vs Usual Care alone (Community-based: mental health and neurological outpatient dementia services) |
UK | RCT CEA+CUA (24 months) |
209 / 209 (140 / 69) |
Health and social care system |
|
|
In the base-case analysis: 1) considering carer-and- patient costs combined, Incremental cost per unit of outcome < 0; 2) considering carer-only costs: £244 per QoL-AD point (patient) £12,400 per QALY gained (caregiver) £179 per HADS point (caregiver) |
High | 1) Considering carer-and- patient costs combined, START dominates usual care when looking at carer outcomes, total HADS score and QALYs (outcomes are better and costs not significantly different) and the intervention had a 70% probability of being cost-effective in terms of carer QALY gain at the NICE’s WTP threshold of £30,000 per QALY. 2) Considering carer-only costs, cost per carer QALY is less than the lower NICE threshold with a 75% likelihood of cost-effectiveness at the NICE’s WTP threshold of £30,000 per QALY. | Medium |
| 1.3 | Livingston et al. 2019 [118] | Individual therapy program for informal caregivers (START) vs Usual Care alone (Community-based: mental health and neurological outpatient dementia services) |
UK | RCT CEA (6 years follow-up) |
222 / 222 (150 / 72) |
Health and social care system |
|
|
Intervention is cost-saving compared to usual care but has positive effects on outcomes (e.g. mean difference in HADS scores of -2.00 points) | High | The positive difference in outcomes is small but statistically significant, greater than the minimally clinically important difference and is sustained after 6 years. The difference in costs appears to be economically large (e.g., cost per patient in the intervention group is around a third of the cost in the control group) although for PwDs there was no significant difference in time to care home admission or death. | Medium |
| 2 | Nichols et al. 2008 [73] | Psychosocial intervention for informal caregivers (REACH II) Vs Usual care (Community-based: home) |
USA | RCT CEA (6 months) |
112 / 112 (55 / 57) |
Societal |
|
|
$4.96 per hour not spent in caregiving (the cost of an additional hour of non-caregiving time that could be “purchased” by the intervention) | High | Intervention was cost-effective if one was willing to spend $4.96 per day for 1 extra hour of non-caregiving time for each family caregiver. Moreover, the intervention could be thought of as being financially positive because it resulted in $10.56 ($8.12 of caregiver hourly wage × 1.3 hours) of time gained versus $4.96 of intervention cost per hour per day per caregiver. | Medium |
| 3 | Drummond et al. 1991 [69] | Caregiver support program (nurse visits, support groups and respite care) vs Usual Care (conventional community nursing care) (Community-based: home) |
Canada | RCT CUA (6 months) |
0 / 42 (22 / 20) |
Health and social care system | — |
|
Mean ICER: CA$20,036 per QALY gained | Moderate | Incremental cost per QALY gained compares favourably with other health care interventions. However, evidence of cost-effectiveness was considered limited due to the statistically non-significant difference in outcome levels. Further larger studies are required. | Low |
| 4 | Shaw et al. 2020 [76] | FamTechCare telehealth intervention to assist caregivers vs Telephone intervention (Community-based: home) |
USA | RCT CEA (3 months) |
56 / 68 (31 / 37) |
Health and social care system | — |
|
Mean ICERs: $222.17 (per dyad) for 1-point improvement in CES-D score (depression); $436.53 (per dyad) for 1-point improvement in SSCQ score (competence) | Moderate | Partial cost-effectiveness: a caregiver’s WTP amount on improvement in SSCQ score (based on a different trial focused on training caregivers) is used as a threshold to determine the cost-effectiveness of the intervention. However, the authors recognise that established external WTP thresholds for the considered units of effectiveness do not exist. | Low |
| 5 | Gaugler et al. 2003 [131] | Adult day care service to support informal caregivers vs Usual Care (Community-based: Day Care) |
USA | Prospective study CEA (3 months; 1 year) |
0 /201 (80/121) |
Societal |
|
|
Mean ICERs calculated as the cost necessary to alleviate role overload and depression by one unit: 1) $6.83/day per unit of ROS score; $2.90/day per unit of CES-D score (over 3-months period); 2) $4.51/day per unit of ROS score; $2.20/day per unit of CES-D score (over 1-year period) | Moderate | Partial cost-effectiveness: the daily costs of carer’s benefits were reduced over a 1-year period. Long-term utilization helped to lessen the time carers spent managing symptoms associated with dementia (i.e., ADL dependencies and behaviour problems) and allowed caregivers to spend more time in work-related activities. No established external WTP thresholds for unit of effectiveness. | Low |
| 6 | Joling et al. 2013 [72] | Family meetings for informal caregivers vs Usual Care (Community-based: home) |
Netherlands | RCT CUA (12 months) |
192 / 192 (96 / 96) |
Societal |
|
|
Mean ICERs: -€807,703 per QALY (dyad: carer +patient), -240,247 per QALY (patient), -€24,472 per QALY (caregiver) [intervention more costly and less effective] |
Low/ Absent |
The intervention is not considered cost-effective. Since the differences in effects on all outcomes were very small, this resulted in very large ICERs that are very sensitive to uncertainty in incremental effect. The probability that the intervention was considered cost-effective was 36% for the outcome QALY per dyad (patient+carer) when the ceiling ratio is set at €30,000/QALY). For caregivers separately this probability was 85% for a ceiling ratio of €30,000/QALY. For patients this probability was around 29% for a ceiling ratio of €30,000/QALY. | High |
| 7 | Wilson et al. 2009 [132] | Social care intervention for informal caregivers (contact with a befriender facilitator) vs Usual Care (Community-based: home) |
UK | RCT CUA (15 months) |
0 / 190 (93 / 97) |
Societal | — |
|
Mean ICERs in the base-case: £105,954 per QALY (caregiver) £28,848 per QALY (carer+patient) | Low/ Absent |
It is unlikely that befriending is a cost-effective intervention. The intervention had only a 42.2% probability of being cost-effective in terms of carer QALY gain at the NICE’s WTP threshold of £30,000 per QALY. The intervention had only a 51.4% probability of being cost-effective in terms of dyad (carer+patient) QALY gain at the NICE’s WTP threshold of £30,000 per QALY. | High |
ADLs: Activities of Daily living; CES-D: Center for Epidemiologic Studies Depression Scale; COPE: self-completed measure of carer coping strategies; CQLI: Caregiver Quality of Life Instrument; CSRI: Client Service Receipt Inventory; CUA: Cost-utility analysis; EQ-5D: EuroQol-5 Dimensions; HADS: Hospital Anxiety and Depression Scale; IADLs: Instrumental Activities of Daily Living; MMSE: Mini-Mental State Examination; MINI: Mini International Neuropsychiatric Interview; NPI: Neuropsychiatric Inventory; QALYs: Quality Adjusted Life Years; RCT: Randomised controlled trial; RMBPC: Revised Memory and Behavior Problem Checklist; ROS: Role Overload Scale; SF-12: Short Form questionnaire-12 items; SSCQ: Short Sense of Competence Questionnaire; STAI: State-Trait Anxiety Inventory; ZBI: Zarit Burden Interview (self-reported questionnaire used to assess carer burden).