To test the hypothesis that prophylactic granulocyte macrophage–colony-stimulating factor (GM-CSF) decreases invasive fungal disease (IFD) in patients with neutropenia receiving allogeneic hematopoietic cell transplantation (HCT)

Patients were randomly assigned to receive GM-CSF, GCSF, or a combination of GM-CSF and GCSF. Administration was begun on day 5 after transplantation and continued until neutrophil recovery (ANC > 1.5 x 10⁹ for two days). If ANC declined within five days after stopping the CSF, CSF was resumed until neutrophil recovery again. All received antibiotic prophylaxis with levofloxacin and antifungal prophylaxis with oral fluconazole. Patients were followed for the study for 100 days post transplantation.

• N = 183   
• MEAN AGE = 31.76 years
• AGE RANGE = 13–60 years
• MALES: 72.1%, FEMALES: 27.9%

• SITE: Multi-site    
• LOCATION: China

PHASE OF CARE: Transition phase after active treatment

• Randomized, prospective, open-label, three-group trial

• IFD was determined as proven, probable, or possible according to the European Organization for Research and Treatment of Cancer (EORTC) and the National Institute of Allergy and Infectious Diseases (NIAID) Myoses Study Group (MSG)
• Invasive Aspergillus was diagnosed according to the Infectious Diseases of Society of America's guidelines

No significant differences existed between groups in the prevalence of proven and probable IFD from molds or yeasts overall. In the G-CSF only group, 11.6% had IFD attributable death, compared to one patient in each of the other study groups (p = 0.008). In multivariate analysis to include potential confounders, risk of death was 4.496 times higher (95% confidence interval [CI] [2.5, 8.1]) in patients with proven or probable IFD compared to those without IFD. Those receiving only G-CSF had a significantly higher 100-day mortality rate (p = 0.037). All infection-related mortality was lowest in the GM-CSF group (p = 0.011).

The findings showed that GM-CSF was more effective than G-CSF in the prevention of infection, fungal disease, and infection-related mortality at 100 days in patients undergoing allogeneic hematopoietic cell transplantation (HCT).

• Risk of bias (no blinding)

This study suggests that the use of GM-CSF versus G-CSF is more effective for infection prevention in patients undergoing allogeneic HCT with neutropenia. The effective use of colony-stimulating factors has been shown to reduce infection and IFD-related mortality in at-risk patients.

To examine effects of live music therapy on quality-of-life indicators, medications administered, and length of stay in patients undergoing brain surgery

Patients were met 30–45 minutes prior to surgery in the outpatient surgery check-in area, inpatient room, or preoperative holding area and completed baseline study measures. Patients in the experimental group received 20–30 minutes of patient-preferred live music and completed postintervention measures prior to surgery. Those in the experimental group received the music intervention each subsequent day of hospital stay, and completed both pre- and postintervention measures. Patients, family members, and visitors could participate by singing, playing rhythm instruments, or listening. Techniques included lyric analysis, songwriting, progressive muscle relaxation, and guided imagery. Control group patients also completed study measures postoperatively and daily during their hospital stay.

• The study reported on a sample of 27 patients.
• Median patient age was 48 years, with a range of 8–73 years.
• The sample was 55.6% female and 44.4% male.
• Diagnoses included aneurysm, meningioma, neuralgia, malignant neoplasm, osteoma, angioma, intracranial abscess, and metastatic cancer to the brain.
• Surgical procedures varied according to patient situation.

• Single site
• Inpatient setting
• Florida

Patients were undergoing the active treatment phase of care.

A randomized controlled trial design was used.

• Visual analog scale for mood, pain, relaxation, stress, and perception of hospitalization
• Total milligrams of vicodin and morphine
• Medications for nausea
• Hospital length of stay

There were no significant differences between groups for anxiety, mood, pain, perception of hospitalization, relaxation, or stress. There were no differences between groups for medications used. There was no significant difference between groups for length of stay.

Results do not support an effect of live music therapy on anxiety, pain, medication use, or length of hospital stay in patients undergoing brain surgery.

• The study had a small sample, with less than 30 patients.
• The study design lacked an attentional control.
• The practicality of providing such an intervention preoperatively in a check-in area or preoperative holding area is questionable, and the authors did state that many of these interventions were interrupted. Sessions during hospital stay were also interrupted for various aspects of care.
• Range of age was very broad, and no other demographic information about patients was provided.

This study does not demonstrate effectiveness of music therapy in hospitalized patients undergoing brain surgery. Practical application of this type of intervention in most acute inpatient settings and perioperative settings is questionable.

To evaluate the success of additional professional team member weekly home visits, beyond support provided via specialist palliative care, on carer distress, burden, quality of life, satisfaction with care, and bereavement outcome

Carers were randomly assigned to either a control group (usual care = specialist palliative care team help and support in the home and clinic) or an intervention group (usual care plus a trained carer advisor who privately met with the carer to deliver advice and support outside the home and address domains of carer need each week over a six-week period). Carers completed mailed questionnaires (three instruments at 4, 9, and 12 weeks after randomization to group). Brief, semistructured interviews with carers at the final assessment time provided information about acceptability and satisfaction with the intervention.

• The sample was comprised of 271 carers (80% female and 20% male).
• Mean age of carers was 56.3 years (range = 16–92 years).
• All patients of carers received regular care at one of three cancer networks employing seven specialist palliative care teams.
• Carers scored above a threshold of 5/6 on the General Health Questionnaire (GHQ-28).
• Most carers were white (86%), and 64% were married or were partners with the patient.
• Approximately 30% of carers had a college education.

• Multisite/other setting
• London, England

An experimental design with generalized linear latent and mixed models (GLLAMM) approaches was used, as well as repeated measures with a brief intervention.

• General Health Quesionniare–28 (GHS-28): Assessed carer psychological distress*
• Carer Strain Index: Assessed baseline carer burden and at 4, 9, and 12 weeks*
• Carer Quality of Life Index: Assessed baseline carer quality of life and at 4, 9, and 12 weeks*
• Eastern Cooperative Oncology Group (ECOG): Assessed baseline patient physical performance status

* No reliability or validity data were given in the article with study use; references appear to address this area.

About 30% of carers in both the control and intervention groups decreased their GHS-28 scores to show less stress at each assessment point in the study. Mean GHS-28 scores dropped at 4- and 9-week assessment times but then increased by the 12-week assessment. The intervention group appeared to experience greater but statistically nonsignificant improvement in GHS-28 scores as compared to the usual care control group. GLLAMM, used to more specifically analyze the influence of the intervention on GHS scores, did not show any significant interaction effects of time and treatment. By the end of the study, 40% of patients had died. Carers noted the emotional support provided by the trained advisor as most beneficial, 20% noted that the intervention came “too late” to help, and almost 30% noted that more advisor sessions would have been helpful.

This study offers insight into the difficulties of collecting data on an intervention with carers who assume responsibility for a patient receiving palliative care due to a diagnosis of cancer. The fact that more 60% of carers scored above the threshold on the GHS-28 at baseline indicates that many carers show strain and would benefit from professional help during the cancer end-of-life journey. Results of this study did not show significant effects of the intervention, although a percentage of subjects identified that the intervention was helpful. It is not clear how different this intervention was from the usual care, which was provided by clinicians specialized in palliative care.

It is not clear whether the three instruments used in this study accurately assessed the variables of interest due to absence of information on the instruments in the article. For example, the authors wished to measure burden but used an instrument to measure strain. One must ask if these terms are conceptually equivalent to support use of the Carer Strain instrument to meet the aims of this study.

It is not clear whether each carer in the intervention received a tailored six-week program or whether all carers in that group received “all domains of care” (p. 143). It also is not clear how the intervention changed when it was delivered outside the home and perhaps in a carer’s workplace where distractions and lower privacy might exist (influences on external validity of study). The lack of specificity about differences between usual care (specialist palliative care teams) and the carer advisor intervention leads one to understand study findings of no significant effect with the brief intervention. One wonders if the use of ECOG scores obtained on patients could have predicted inclusion of carers who would have had a greater chance of concluding the study with a viable family member.

Additional investigation of effective interdisciplinary interventions to improve the quality of life of carers engaged with end-of-life care must be completed to uncover needs of carers during that vulnerable time, and to determine the most appropriate timing of such interventions This study indicated that carer quality of life deteriorated over the 12 weeks of the study despite a professionally trained carer advisor. The authors added valuable information about ways to refine their intervention to be useful in future studies. Continued search for evidence-based components of an intervention, optimal frequency and intensity (as well as sites for delivering it), and assessments throughout the intervention to determine its effectiveness will help support improved care for carers who commit to others despite their own grief.

To test the effects of art-making classes to reduce anxiety and stress among caregivers of patients with cancer

Art-making classes were offered as one part of an already established art program. The class involved with the research began with discussion of the study. Study participants completed self-report instruments and provided a saliva sample for cortisol testing. The art-making class was given over a two-hour period, and repeat questionnaires and saliva testing were done at the end of the session. Classes were delivered twice weekly by volunteer art interventionists in a residential facility. A variety of art-making projects were used in classes. Research team members attended each class and documented field notes during each session. Interventionists were trained in processes of caregiver experiences based on the end-of-life phase of experiential theory.

• The sample was comprised of 69 caregivers (80% female, 20% male).
• Mean caregiver age was 48 ± 14.47 (range = 18–81 years).
• Disease types of patients were not stated.
• Participants included Hispanics, Caucasians, Carribean Islanders, and individuals from other cultures.
• Of the sample, 75% were the primary caregiver of the patient, 41% had provided care for six months to one year, and 56% had high school formal education or less.

• Single site
• Other setting
• Miami, Florida, United States

• Late effects and survivorship
• Palliative care

A pretest/post-test quasi-experimental design was used.

• Beck Anxiety Inventory
• Salivary cortisol
• Field notes of participant comments

Anxiety measures showed a significant reduction in scores of the Beck Anxiety Inventory after the session, with preintervention of 7.28 ± 6.8 and postscore of 2.49 ± 4.5 (p < 0.01). No significant changes in cortisol level were reported. Field notes indicated that participants shared efforts, offered suggestions to each other, and became better acquainted. Numerous subjects refused to give samples for salivary cortisol.

Art-making classes appeared to produce a short-term reduction in anxiety level among caregivers of patients with cancer.

• The sample was small, with less than 100 participants.
• Risk of bias existed due to no control group, no binding, no random assignment, and no appropriate attentional control condition.
• Measurement validity and reliability are questionable.*
• Other/*explanation: Cortisol levels can be expected to vary according to time of day. No information is available about the time of specimen collection in the study, and it is not known whether all patients had art-making sessions at the same time of day. Pre- and postanxiety inventory measures showed high variability, suggesting that mean scores may not be representative of the group. No information is available about how many sessions people attended. Sessions were also attended by individuals who were not part of the study or who had refused to provide consent for participation. Although numerous subjects refused to give salivary samples, the authors did not say how many refused or discuss relevant missing data. The authors noted lack of funds for creative approaches used. It is not clear if changes in anxiety were truly due to the use of art in these sessions, or the support group type of interactions that occurred among participants.

Findings suggest that participation in art making may reduce anxiety among caregivers momentarily, and group participation can provide an avenue for supportive caregiver interactions.

To test hypotheses that family caregivers would experience reduced stress and anxiety and have increased positive emotions from an art-making intervention

Art-making supplies were taken to patients’ bedsides or to the outpatient chemotherapy site to show patients and caregivers items that could be made. Caregivers decided on one or more activities that they could do with or without the patients’ involvement. Caregivers were given supplies and shown how to complete the activity. The artist–nurse intervention team then left the area and returned to monitor progress and offer assistance every 15–30 minutes. Participants completed study questionnaires prior to and immediately after the intervention.

• The sample was comprised of 40 family caregivers.
• Mean caregiver age was 51.43 ± 15.38 years.
• Of the sample, 78% were the primary caregivers for the patients, 75% were women, and most were spouses.
• Most caregivers had provided care for six months or less.

• Single site 
• Multiple settings
• Florida, United States

Mutliple phases of care

A pretest/post-test quasi-experimental design was used.

• Mini Profile of Mood States (miniPOMS)
• Beck Anxiety Inventory
• Derogatis Affects Balance Scale

The presession stress score mean was 13.27 ± 6, and the postscore was 9.85 ± 5.84 (p < 0.001). Cohen’s d calculation on stress scores was d = 0.44, suggesting a large effect size. Postintervention anxiety scores declined but were not reported to be statistically significant. Significantly more positive emotions were reported in the post-test evaluation  (p < 0.001). It was noted that individuals who participated in the hospital inpatient units had multiple interruptions.

Involvement in art making was associated with reduction in stress and increased positive emotions immediately after the involvement. Participation at the bedside in the inpatient area was complicated by multiple interruptions.

• The sample was small, with less than 100 participants.
• Risk of bias existed due to no control group, no binding, no random assignment, and no appropriate attentional control condition.
• Whether the nature of art making itself, or any type of distracting activity, was responsible for the changes seen is not clear.

Involvement in art making may be helpful for short-term stress reduction in caregivers of patients with cancer. Further well-designed research in this area is needed to evaluate this approach.

The purpose of the study was to demonstrate bioequivalence of two different filgrastim products.

Patients were randomized to receive one of the two types of filgrastim at the same dose and schedule. Treatment was 5 mcg subcutaneously daily on day 2 of chemotherapy in each cycle, and continued until absolute neutrophil count (ANC) was greater than 3 x 10⁹/L or treatment had been given for 14 days.

• 276 participants were studied.
• Mean age was 49.8 years (SD = 8.88)
• All were female
• All had breast cancer at various stages and were receiving chemotherapy with doxorubicin and docetaxel.
• Patients did not have any radiotherapy within six weeks of study entry, or prior chemotherapy within four weeks.
   

37 European outpatient centers in various countries

Mutliple phases of care

Randomized, double-blind phase III

• Duration of severe neutropenia for each chemotherapy cycle defined as ANC less than 0.5 x 10⁹/L
• Time to ANC recovery (greater than 3 x 10⁹/L)
• Incidence of febrile neutropenia (ANC less than 0.5 x 10⁹/L and temp 38.5ºC or higher)
• Incidence of documented infection
• NCI CTCAE [v.3.0]
   

Incidence of severe neutropenia was 77.6% in one group and 68.2% in the other, with no statistically significant difference. Duration of severe neutropenia across groups in cycle 1 ranged from 1.3 –1.6 days on average, and was lower in both groups in subsequent cycles. There were no differences in outcomes between the two. Those receiving Hospira filgrastim had a slightly higher incidence of  bone pain than Amgen filgrastim; however, overall prevalence of skeletal pain was similar in both groups.

 The results of this study showed that these two different preparations of filgrastim are bioevquivalent.

• Risk of bias (sample characteristics)
• The sample was almost 98% Caucasians. 
• Those receiving Hospira filgrastim who had more bone pain, also had a higher prevalence of bone metastases at baseline. It is not clear if these may have been related.

 This study was designed purely to demonstrate bioequivalence of these two filgrastim products.

To assess the safety and efficacy of long-term repeated dosing of osmotic extended-release oral delivery system (OROS) hydromorphone used to relieve chronic pain

A patient who entered this study after completing a comparative dose-conversion study with OROS hydromorphone continued taking OROS hydromorphone at his or her stable dose. This study also included patients who had participated in a comparison of hydromorphone immediate release (IR) and OROS hydromorphone. These patients began the OROS hydromorphone study by taking 50%–100% of their established dose; dose adjustments were allowed after a minimum of two days. Dose adjustments were usually in 8 mg increments. The target duration of treatment was at least one year. Adverse events were assessed monthly, and physical exams were conducted and vital signs assessed every three months.

• Of all patients in the study, 106 completed the study (388 had enrolled, and 72.7% withdrew). The study sponsor terminated 80 patients (28.4%) from the study because of a decreasing supply of study medication.
• Mean patient age was 50 years. Age range was 27–91 years.
• Of all patients, 50.8% were female and 49.2% were male.
• Patients in the study were adults, with chronic cancer pain or chronic pain unrelated to malignancies, who were receiving stable doses of OROS hydromorphone equal to at least 8 mg/day. Approximately 20% of the initial sample had cancer-related pain.

• Multisite
• Outpatient
• 56 centers in the United States and Canada

Multicenter open-label extension trial

• Brief Pain Inventory (BPI), patient ratings on a 0–10 scale (0 = no pain, 10 = worst pain imaginable)
• Pain relief as measured by the patient on a 0%–100% scale
• Scale measuring extent to which pain interfered with physical activity or social functioning (0 = no interference, 10 = complete interference)
• Global ratings, by patient and investigator, of overall medication effectiveness (1 = poor, 2 = fair, 3 = good, 4 = very good, 5 = excellent

BPI ratings of worst pain, least pain, and average pain were essentially stable throughout the study. Median daily dose of study medication increased from 32 mg at baseline to 40 mg at month 3 and 48 mg at months 6, 9, and 12. The most frequently reported adverse events were nausea (which 24% of patients experienced) and constipation (which 19.3% of patients experienced). The side-effect profile was similar to that of other sustained-release opioids. Most side effects usually resolved over time, although constipation was did not resolve. Laxatives can manage constipation effectively.

Authors concluded that the benefits of OROS hydromorphone were maintained when daily administration was continued. Once-daily OROS hydromorphone appeared to be safe and effective in controlling moderate to severe chronic pain.

• Only 20% of patients in the study had cancer-related pain upon study entry. Authors did not specify the number of patients with cancer-related pain who completed the study and were included in analysis. Given the relatively small number of patients with cancer-related pain, the degree to which overall findings are generalizable to the oncology patient population is unclear.
• The study had risks of bias due to no blinding and no comparison group. 
• Authors did not discuss rescue medication or breakthrough pain.

Of patients who entered the study, 52.3% experienced a treatment-related event. (The side effects experienced by anyone who entered the study and who took at least one dose of OROS hydromorphone were reported.) Therefore, the tolerability of OROS hydromorphone, used long term, appears limited. The single dose required for pain management may be advantageous for patients who must consume multiple oral medications.

To assess the efficacy and tolerability of oral once-daily extended-release hydromorphone in the treatment of chronic cancer pain

The study period consisted of three phases. In the first, study participants were stabilized on previous opioid therapy. This phase lasted for at least three consecutive days on which the total daily dose remained unchanged and fewer than four rescue doses were taken. Visit 1 occurred in this phase. The second phase consisted of conversion to once-daily extended-release hydromorphone. Dose conversion was 5:1 morphine to hydromorphone, with a minimum starting dose of 8 mg hydromorphone. In cases of conversion from transdermal fentanyl, initial hydromorphone was 8 mg for each 25 mg/hour fentanyl. Phase 3 consisted of hydromorphone titration over 3–21 days Visit 2 occurred during phase 3. Each patient's dose was titrated in increments of 25%–100% of current hydromorphone dose. If not stabilized after 21 days, patients were discontinued from the study. Participants who achieved stabilization for 14 days were in maintenance phase (visit 3) and were treated on an outpatient basis through five study visits. Visit 4 was at the midpoint of the maintenance phase. Visit 5 was at the end of the phase. Immediate-release hydromorphone was available for the treatment of breakthrough pain.

The number of enrollees was 148. Of the enrollees, 127 patients received the study medication and 67% completed the study. Patients received at least 45 mg morphine and had stable analgesic requirements.

• Multicenter
• 30 sites throughout the United States and Canada

Open-label, repeated-dose, single-treatment study

• Short Form of the Brief Pain Inventory (BPI)
• Adverse events
• Scale comprising rankings of worst, least, and average, to measure pain intensity
• Scale, 0–10, to measure pain
• Scale, 0%–100%, to measure pain relief
• Scale, 0–10, to measure the extent to which pain interfered with activity, mood, walking, work, relationships, sleep, and enjoyment of life
• Scale, comprising five points, to measure patients' and investigators' perceptions of medication's overall effectiveness

Dose stabilization occurred in 119 of 127 patients (94%); 77% of patients achieved stabilization without titration. Mean BPI pain intensity and pain interference scores decreased significantly. Mean pain relief level remained stable after conversion and throughout treatment. Adverse events were as expected. Authors noted no clinically significant changes in vital signs. Of all patients, 87% received hydromorphone for seven days or longer; 40%, for 25 days or longer. Of all patients, 94% achieved dose stabilization. Mean time to dose stabilization was 3.6 days. Dose increased 38% from initiation to stabilization but decreased slightly at the end of the maintenance phase. Frequency of rescue medication declined. Ratings indicated a decrease in pain, but the decrease of pain intensity on average was the only significant pain-related statistic (p < 0.001). Decreases in pain intensity were accompanied by a significant decrease in pain interference scores for all categories (p < 0.05). Of all patients, 83% experienced adverse events (nausea, constipation, vomiting, diarrhea, and somnolence). Serious adverse events occurred in 20 patients (16%). Authors reported that only one of these serious adverse events, confusion accompanied by hallucinations, was due to hydromorphone. Four deaths occurred during the study, all due to disease, not treatment.

Patients with chronic cancer pain can easily undergo conversion from previous opioid to stabilization on once-daily oral extended-release hydromorphone. Authors noted that use of this form of hydromorphone offered acceptable clinical efficacy and safety and the convenience of once-daily dosing. Authors concluded that the 5:1 ratio conversion was effective and that the conversion was not a problem for most patients. Adverse events were consistent with those expected with the use of other opiods.

• The study had a risk of bias due to the open-label design and lack of control group.
• Patients’ previous opioid use was not standardized.
• The pharmaceutical and medical systems company Alza sponsored the study. Some investigators had conflicts of interest due to affiliation with pharmaceutical companies.

The study compared acupuncture tovenlafaxine for 12 weeks with health measurements for one year.

The enrolled 50 women, with 25 randomized to each arm. 

Inclusion criteria:

• Stage 0–III pre- or postmenopausal patients with breast cancer on hormone therapy with tamoxifen or arimidex
• 14 hot flashes per week
• 18 years of age
• May have been treated locally with surgery or radiation and must have completed chemotherapy
• May be receiving radiation therapy but otherwise must be within five years after treatment
• Must be on a stable dose of hormone therapy for four weeks or more without plans to discontinue therapy for the duration of the study
• Karnofsky performance status
• Life expectancy of at least six months

This was a randomized, controlled trial.

Participants completed:

• Hot Flash Diary (number and severity of hot flashes
• Menopause- Specific Quality of Life Questionnaire (MenQOL)

By two weeks after treatment, the venlafaxine group experienced significant decreases in hot flashes, and hot flashes in the acupuncture group remained at low levels. The venlafaxine group experienced 18 incidences of adverse effects (e.g., nausea, dry mouth, dizziness, anxiety), whereas the acupuncture group experienced no negative adverse effects. Acupuncture had the additional benefit of increased sex drive in some women, and most reported an improvement in their energy, clarity of thought, and sense of well-being.

Both groups exhibited significant decreases in hot flashes, depressive symptoms, and other QOL symptoms. Acupuncture was as effective as venlafaxine.

The study was limited by its small sample size.

To determine, by using a systematic review, which, if any, treatments have been found to be effective in reducing depression in patients with lung cancer

• Authors consulted these databases: MEDLINE (1948–Oct. 1, 2011), EMBASE Classic and EMBASE (1947–week 41, 2011), PsycINFO (1806–week 2, 2011), CINAHL Plus (1937–October 2011), and the Cochrane Central Register of Controlled Trials. Search keywords were depression, lung cancer, treatment, and systematic review.
• In this randomized controlled trial, adults were 18 years old or older and had received a definite diagnosis of lung cancer.
• The trial evaluated the efficacy or effectiveness of a pharmacologic or nonpharmacologic intervention intended to improve patients’ symptoms or quality of life. Depression outcomes were assessed using a standardized measure.
• Excluded from the analysis were nonprimary publications for which the full paper could not be obtained for data extraction.

The total number of references retrieved was 143. The evaluation method consisted of the review, by two independent researchers, of full articles.

• The final number of studies included was six.
• The sample range, across studies, was 64–549.
• Patients in the largest sample were newly diagnosed or new to the clinic; those in the smallest sample had a prognosis of 3–12 months. Most were from ambulatory clinics. Some patients had early-stage disease and some had metastatic disease.
   

• Phase of care: multiple
• Application: eldercare

No trials aimed to evaluate the effectiveness of treatments for depression in people with lung cancer. The six trials in the sample discussed interventions intended to improve symptoms related to quality of life, and each trial included a measure of depression as a secondary measure. The interventions, depression measures, and time of measurement varied. The interventions included breathlessness advice and discussion, education about self-referral for local psychosocial resources, counseling, coping skills training (including progressive muscle relaxation and symptom management strategies), early introduction of palliative care, and supportive psychotherapy. Studies indicated that enhanced care was more effective in reducing depression symptoms than was standard care.

Patients with lung cancer tend to be older adults with medical comorbidities, and these patients tend to suffer severe physical deterioration. Although standard depression treatments may be a reasonable course for treating depressed people with lung cancer, no randomized controlled trials (RCTs) guide clinicians in treating this population.

• The review was not limited to interventions designed specifically to treat depression.
• Participants were not necessarily depressed at the time of trial enrollment.
• Reviewers found that reports of the procedures for participant recruitment and the collection of outcome data frequently lacked detail.

No evidence guides clinicians who are caring for this specific population; well-conducted RCTs are urgently needed. Analysis indicates that clinicians may consider, as tools to reduce depression, depression treatments effective in older adults in the general population and those with medical comorbidities.