Use of comprehensive geriatric assessment in general practice: Results from the ‘Senta Pua’ project in Brazil

Abstract

Background: General practitioners (GPs) are responsible for most of elderly consultations in Brazil. Being aware of the needs of older adults is important for these professionals. A comprehensive geriatric assessment (CGA) is an important tool for assessing the elderly. Objective: The aim of this study was to evaluate (a) whether a CGA can identify previously unknown health problems, and (b) whether a CGA is accepted by elderly patients. Methods: We conducted a cross-sectional study by inviting all outpatients of the Sao Paulo Air Force Hospital aged 60 years and older to participate. Consenting patients were examined using a CGA. Health problems revealed by the assessment were compared to those previously known from the patients' charts. Patients' acceptability of the CGA was evaluated by a short questionnaire. Results: 170 patients were included in the final analysis. Mean age was 75.7 years (SD: 8.0 years), 65.8% was female. On average, patients had 3.2 (SD: 1.9) geriatric problems gleaned from their charts and 9.0 (SD: 3.2) revealed by CGA. The most common health problems disclosed by the CGA were pneumococcal vaccination more than 3 years earlier, followed by sleep disorder, sedentarism, urinary incontinence, fall risk and cognitive impairment. The mean duration of CGA was 52 min and patient acceptability was high.

Conclusion: The present study using the CGA in a population of patients aged 60 and above disclosed, on average, 6 medical problems previously unknown to the GP. The acceptability by patients was good.

Key words::

• Geriatric assessment
• aged
• general practice
• geriatrics

Introduction

According to the Brazilian Geriatrics and Gerontology Society, there are 600 geriatric specialists for 3 million elderly aged older than 80 years in Brazil, equivalent to one geriatric specialist for every 5000 older adults (1). Compared to the United States (approximately 6 geriatricians per 5000 80+ year-olds) and a European country like the UK (2 per 5000 80+ year-olds) this is a low number (2,3).

Results of a recent survey show that approximately 24% of Brazil's 330 000 physicians practice in primary care settings (4). Thus, general practitioners are responsible for most of consultations in the elderly in Brazil. Therefore, awareness of the medical needs of this group of the population is important for these professionals. A ‘comprehensive geriatric assessment’ (CGA) might serve as a useful tool for obtaining a systematic and comprehensive overview of relevant health problems in old age.

In Brazil, GPs are considered generalist physicians. Their role embraces that of Family Physicians (residency in family medicine), Primary care Physicians (those that work in primary care, even without a residency) and specialists in Internal Medicine. However, these physicians usually have limited training in geriatric medicine or use of CGAs.

The concept of CGA was first introduced by Marjorie Warren in 1936 and differs from a standard medical evaluation by including nonmedical domains, emphasizing functional ability and quality of life, and involving interdisciplinary teams. CGAs may be helpful to reveal health-related problems, create reasonable treatment plans through optimal use of health care resources, and facilitate coordination of care, including determining the need for and the site of long-term care (5).

Use of CGA has been extensively studied, having proven benefits for survival and functionality among hospitalized patients (6). In outpatients, results have been conflicting, ranging from improved functional status (7) to no changes in survival (8). Some authors believe CGAs provide a sound basis for developing a reasonable treatment plan that takes into account all relevant factors. Nevertheless, one of the main features of outpatient CGA could be the ability to detect problems unidentified by general practice (9,10).

Several studies have assessed the implementation of CGA in General Practice (9–11), but to our knowledge no such studies have yet been conducted in Latin America. The aim of this study was to evaluate the applicability of a CGA, i.e. whether a CGA can identify previously unknown health problems, and whether a CGA is accepted by elderly patients.

Methods

Study design

A cross-sectional study was carried out at the São Paulo Air Force Hospital (SPAFH) in Brazil between March 2007 and December 2007. The Research Ethics Committee of the São Paulo Air Force Hospital approved the present study.

Location

The SPAFH is a reference hospital for the Brazilian air force in the state of São Paulo (region in South Eastern Brazil with approximately 40 million inhabitants) and treats military personnel and their relatives. In comparison to the Brazilian general population, the study participants had a higher educational and socio-economic level, and the majority was Caucasian.

Sample selection

SPAFH provides approximately 250 primary care consultations per week. However, the hospital has neither a Geriatrics department nor a geriatrician. Due to the high number of elderly served by the hospital, we decided to introduce a specific assessment for evaluating the aged. For this purpose, we created a standardized consultation based on a CGA for application by the general practitioner. Notices about the CGA were displayed in prominent locations around the hospital such as the reception, cafeteria, hallways, and laboratories. Furthermore, flyers were handed out to patients attending the outpatient clinic. These flyers contained the following information: ‘a geriatric assessment is being conducted by our doctors every Monday, Wednesday and Friday. If you would like to participate, please register at the front desk.’

Patient participation in the study was voluntary and study entry was by patient application with consecutive selection. After fully explaining the procedures and obtaining written informed consent from the patient, physicians (general practitioners) conducted consultations using a CGA. To allow time for completion of the consultation and the study protocols, one consultation per hour was scheduled.

Participants

All patients recruited were aged 60 years or older who agreed to participate in the study. Patients with cognitive impairment precluding completion of the test (advanced dementia or inability to communicate) or aged younger than 60 years were excluded.

Screening instrument

The screening instrument was based on the CGA recommended by the Brazilian Geriatrics and Gerontology Society (12), and modified by literature review (9,10,13). The areas covered by CGA and screening instruments are shown in Table I.

Table I. Comprehensive Geriatric Assessment (CGA).

Problem	Diagnostic test (Ref.)	Cut-off
Visual acuity	Jaeger test (29)	≤20/40
Hearing acuity	Whisper test (30)	Failure to repeat 3 whispered numbers
Urinary incontinence	ICIQ-SF (31)	4 questions: ≥ 1 question answered positive
Depression	Geriatric depression scale—15 items (32)	15 questions: ≥ 6 questions answered positive
Cognitive impairment	Mini-mental state examination (33)	30 questions: ≤26 questions positive for high education; ≤18 for low education
Basic and instrumental activities of daily living	Katz scale (34)	6 questions: 0–2 total dependence; 3–4 partial dependence; 5–6 independence
Sleep disorder	Do you suffer from frequent sleeping problems?	Yes/no
Influenza and pneumococcal immunization	Influenza vac. within a one-year period (35)Pneumococcal vac. within a 3 year period (36)	Yes/no (self-report)Yes/no (Self-report)
Fall risk	History of falls	Fall within last 6 months
Cardiovascular risk assessment	Lipid profile (mg/dl) (37)	Chol > 200mg/dl; LDL > 130 mg/dl; HDL <40mg/dl; TG > 150 mg/dl
	Blood pressure (mm Hg)	BP syst > 140 mmHg; BP diast > 90 mmHg
	Hyperglycaemia (mg/dl)	Fasting blood glucose > 126 mg/dl
Medication history	Number of prescribed drugs (38)	> 5 prescribed drugs
Hospital stay	Hospital stay within a 1-year period	Yes/no
Nutrition	Body mass index (39)	BMI normal range: 22–27
Constipation	Rome II diagnostic criteria (40)	7 criteria: ≤2 criteria positive
Osteoarthritis	Physical examination and X-ray for suspect cases	Suggestive X-ray^a
Pain	Presence of pain	Yes/No
	Pain grade	Grade 0–10
Physical activity	Do you have physical activity ≥ 3 times/week?	Yes/No

^aJoint space narrowing, subchondral sclerosis (increased bony formation around the joint), subchondral cyst formation, and osteophytes.

Chart analysis

Prior to the consultations, Patients' charts were examined by a researcher who collected information concerning number and types of diseases recorded, together with number and types of medications. After the consultations, the physician filled out standardized protocols. Between 24 and 48 h after the consultation, data from CGAs were compared with information held in Patients' charts to disclose new diagnoses and changes in the treatment plan as well as identify further medical interventions recommended because of the CGA.

Patient acceptability

After the application of CGA, patients answered a dichotomous (yes/no) self-reported questionnaire that evaluated their perception of CGA. The aspects covered were: (1) whether they were satisfied with the consultation; (2) whether the length of consultation was adequate; (3) whether CGA covered the main aspects; (4) whether they obtained new information on their health status; (5) whether they were embarrassed to answer the questions. A nurse collected the completed questionnaires for analysis.

Statistical analysis

Data were tabulated in Excel for Windows. Statistical analysis was performed using SPSS (version 15.0) and Epi Info for Windows (version 3.6.1.). Descriptive statistics determining means and standard deviation (SD) were used for quantitative variables, whereas simple and relative frequencies were employed for categorical variables. The Wilcoxon-Mann-Whitney test was used to analyse differences between the number of geriatric problems and number of medications, before and after CGA. The level of significance was set at P < 0.05.

Results

Patients

The study was conducted over a 10-month period with sessions three times a week in the mornings (12 h per week). Each day at the clinic was split into four periods (total four hours). This gave a total 492 h available for consultation: 22 took place without prior appointment (all in the beginning of the study), 187 were new consultations, 261 returns (1.39 returns per patient), whilst 22 patients missed their medical consultations.

187 patients were evaluated. Of these, 8 were excluded because they were not physically and cognitively fit to undergo baseline evaluation, 7 were rejected for being younger than 60 years and 2 declined to take part in the study. Thus, the final sample comprised 170 elderly patients (Figure 1).

Figure 1. Study flowchart.

The Patients' main characteristics as recorded in the CGA protocol are shown in Table II. Patients had a mean age of 75.7 (SD: 8.0) years, and were predominantly women (65.8%).

Table II. Patients main characteristics as recorded in the CGA protocol (n = 170).

Variable
Age (mean, SD)	75.7 (8.0)
Female sex (%)	112 (65.8%)
Number of geriatric problems before/after CGA (mean, SD)	3.2 (1.9)/ 9.0 (3.2)	P = 0.000^a
Number of patients with limitations in basic activities of daily living	18 (10.5%)
Hospital stay within a 12-month period	33 (19.4%)
Weight (n = 141)
BMI < 22	20 (15.4%)
22 ≤BMI ≤27	63 (43.5%)
BMI > 27	58 (41.1%)
Cardiovascular risk factors
Hypertension	125 (73.5%)
Diabetes	36 (21.2%)
Hyperlipidemia	74 (43.5%)
Smoking	30 (17.6%)
Number of drugs before/after CGA (mean, SD)	4.0 (2.9)/4.73 (2.6)	P = 0.000^a
Falls within a 6-month period	46 (27.1%)

^aWilcoxon-Mann-Whitney test.

Prevalence of problems

The prevalence of problems is shown in Table III. Some procedures such as measurement of weight (n = 141) or glucose level (n = 129) were not available for all 170 patients. The most common problems found were: pneumococcal vaccination more than three years earlier (98.8%), sedentarism (60.6%), sleep disorders (52.9%), hearing loss (48.8%) and urinary incontinence (46.5%). The medical charts showed a mean of 3.2 geriatric problems versus 9.0 geriatric problems on the CGA, a difference that reached statistical significance (P > 0.001). A similar observation was made for the number of drugs per patient. Prior to the CGA the mean number of prescribed drugs was 4.0, whereas after the CGA the average number of drugs was 4.7 (P > 0.001).

Table III. Prevalence of positive CGA test results, ranked in decreasing order (n = 170).

Problem	Number of people	Prevalence
Pneumococcal vaccination longer than 3 years ago	168	98.8%
Sedentarism	103	60.6%
Sleep disorder	90	52.9%
Hearing loss	83	48.8%
Urinary incontinence	79	46.5%
Hyperlipidemia at assessment	74	43.5%
Overweight or obesity (n = 141)	58	41.1%
Hypertension at assessment	67	39.4%
Last influenza vaccination longer than one year ago	56	32.9%
Low vision	56	32.9%
Fall risk	46	27.1%
Depression (n = 162)	45	26.5%
Osteoarthritis	44	25.9%
Cognitive impairment	40	23.5%
Hyperglycaemia at assessment (n = 129)	17	13.2%
Constipation	16	9.4%

Yield of CGA

Table IV shows the percentage of problems before and after CGA and the percentage of patients identified with the respective problem by the CGA. The most common problems not previously identified but disclosed by CGA were: pneumococcal vaccination more than three years earlier (99.9%), followed by sleep disorder (96.5%), sedentarism (83.4%), urinary incontinence (78.0%), fall risk (77.8%) and cognitive impairment (77.4%).

Table IV. Percentage of patients (n = 170) with problem before and after CGA and percentage of patients with problem identified by CGA.

Problem	Problem known to GP before CGA	Problem known after CGA	Percentage detected by CGA
Pneumococcal vac. > 3 years ago	0.5%	98.8%	99.9%
Sleep disorder	1.8%	52.9%	96.5%
Sedentarism	10.0%	60.6%	83.4%
Urinary incontinence	11.7%	53.3%	78.0%
Fall risk	5.8%	27.1%	77.8%
Cognitive impairment	5.3%	23.5%	77.4%
Hearing loss	11.7%	48.8%	76.0%
Depression	12.3%	32.9%	62.6%
Osteoarthritis	10.0%	25.9%	61.3%
Last influenza vac. > 1 year ago	15.3%	32.9%	53.4%
Low vision	45.9%	78.8%	41.7%
Hyperlipidaemia	26.4%	43.5%	39.3%
Constipation	7.6%	9.4%	19.1%
Diabetes	17.7%	21.2%	16.5%
Hypertension	67.0%	73.5%	8.8%

Patient acceptability of CGA

18 subjects did not return the self-reported questionnaire. Of the 152 patients who completed the CGA, 151 (99.3%) stated they were satisfied with the consultation, 151 (99.3%) stated that the CGA covered the main problems, 123 (80.9%) agreed that they had learned something new from the consultation, 72 (58.5%) reported that they became aware of health problems not previously detected and 51 (41.5%) learned something new about how to deal with their problems. Only 3 (2.0%) patients felt embarrassed by the questions.

The mean consultation time for application of the CGA was 52 min ranging from 30 to 100 min (SD: 11 min). 17 patients (11.2%) stated the consultation was too long, 1 (0.7%) deemed it too short, and 134 (88.2%) considered the time needed was adequate.

Discussion

Main results

On average, patients had 3.2 geriatric problems on their medical charts and 9.0 revealed by the CGA. When considering the percentage of patients identified with a new problem exclusively by CGA, the most common problems were: pneumococcal vaccination more than three years earlier, followed by sleep disorder, sedentarism, urinary incontinence, fall risk and cognitive impairment.

The mean duration of the CGA was 52 min and patient acceptability was high.

Strengths and limitations

The present study has some limitations that should be considered. First, this was a cross-sectional study based on only one consultation, and did not follow up the patients to evaluate dependency, mortality or hospitalization. Second, no control group was used with all participants submitted to CGA. Third, differences between economies, cultures and health care systems may influence the prevalence of problems and thus affect the benefit of CGA in other countries.

In our sample, there was a predominance of females in their mid 70s, in accordance with the findings of other studies conducted in Brazil's elderly population (14,15). Nevertheless, in comparison to Brazil's general population, study participants had a higher educational and socio-economic level, with the majority being Caucasian.

Interpretation of the study results

Pneumococcal vaccination. For pneumococcal vaccination, not all patients over 60 years old should be vaccinated in Brazil (16). Thus, the results found probably do not fully reflect reality because most patients, according to the Brazilian vaccination calendar, were not scheduled to have received this vaccination.

Sleep disorders. Sleep disorder was reported by almost 52.9% of participants where 96.5% of sleep disorder cases were newly identified by CGA. These findings highlight the importance of CGA in identifying new problems often overlooked in general consultations. These results are similar to those of a study carried out by Reid et al. (17), confirming the relevance of sleeping disorders in older age.

Physical activity. Regarding sedentarism, 83.4% of cases were newly identified by CGA. These findings can be explained by the lack of information recorded in the medical charts or by the lack of questioning about this issue. In fact, this knowledge is very important for stratifying cardiovascular risk and for physician counselling. Recently, Kruger et al. evaluated 185 702 older adults and found that 30.0% of this group did not engage in leisure time physical activity. The authors concluded that: ‘Regular physical activity is an important means to maintaining independence, because it substantially reduces the risk for developing many diseases; contributes to healthy bones, muscles, and joints; and can reduce the risk for falling. Health care providers are encouraged to discuss concerns regarding physical activity with their patient’ (18).

Fall risk. Fall risk was newly identified by CGA in 77.8% of cases. Falls and concomitant instability can be markers of poor health and declining function. In older patients, a fall may be a nonspecific sign of several acute illnesses, such as pneumonia or urinary tract infection, or may be a sign of acute exacerbation of a chronic disease (19). Thus, it is important for health professionals that deal with older adults to evaluate this aspect in routine consultations.

Other geriatric problems. Other common geriatric problems such as cognitive impairment, urinary incontinence and hearing loss tend not to be assessed in general practice routine consultations, although the prevalence of these conditions in the elderly is high (9,10).

Medication. Concerning the higher number of drugs as revealed by CGA, we believe that the higher number of prescriptions resulted from the new conditions detected by CGA, including urinary incontinence, cognitive impairment, depression and osteoarthritis.

Patient acceptability

Patients from our study population were satisfied with the CGA consultation and believed it covered all their complaints. These results are in line with other studies (10). The self-reports also revealed that the acceptability of the procedure was high (20). This finding gains importance when considering that the physician-patient relationship can be strengthened because the patient feels confident and understood by their doctor. Several factors could explain these high rates of CGA acceptability by patients: (a) the comfort and possibility of talking without restrictions about their problems; (b) the perception that they were being evaluated as a whole, including psycho-social aspects; (c) the high number of scales used probably boosted confidence in the physician's findings.

Criticisms of CGA

Some authors criticize the CGA because it invariably raises patient morbidity. In our opinion, the CGA allows identification of unrecognized geriatric conditions, as seen in the present study. One clear example is a study conducted by McDowell et al. (21), comparing the difference in recognition of urinary incontinence by the Geriatric Assessment Units (GAU) and physicians in community-based practices (CMD). Recognition rates for UI were significantly higher for GAUs (59.3%) than for CMDs (15.7%). However, in some instances, patients could perceive they have more health problems than previously thought.

Benefits and drawbacks of CGA

Several studies on the CGA have assessed the benefits and drawbacks of this procedure in clinical practice. Studies conducted in hospitalized elderly have shown that CGA was effective in improving survival and function (6,22), reduced the length of the initial hospital stay, and also contributed to reducing direct costs of hospitalized patients (23). However, the studies conducted in outpatients failed to confirm some benefits of the inpatient CGA, reporting no differences in survival (8,24) or hospitalization (25).

Although many studies have been carried out in outpatients, few investigations have been conducted in the General Practice setting. A recent study conducted by Piccoliori et al. (9), found that geriatric screening can detect problems unidentified in general practice (7.8 out of 32 problems), in agreement with our study that identified approximately 9 problems after application of CGA. Mann et al. administered CGA in an outpatient group and concluded that the procedure detected a high number of geriatric problems (6.4 out of 14 problems) (10). Moreover, these authors stated CGA was feasible and well accepted in the general practice sample; findings corroborated by the present study. Another study involving outpatient elderly showed that CGA prevented functional and health-related qualityof-life decline among community-dwelling older persons (26).

Given that use of the CGA could ultimately reduce other specialist consultations, the cost-benefit of the procedure can be positive, as demonstrated in a recent study by Melis et al. (27).

Implications for clinical practice

In our opinion, the main reasons for the overlooking of geriatric conditions are lack of knowledge (including the opinion that many complaints are due to intrinsic conditions affecting the elderly, which do not require treatment, and lack of familiarity of how to treat these conditions) and lack of time.

Concerning lack of knowledge, a Brazilian study concluded that cognitive decline of elderly patients often went undetected by general practitioners (28). According to the author, out of 248 patients, 21 had dementia and 22 were classified as having mild cognitive impairment. However, notes by the general practitioner on cognitive decline appeared in only seven (16.27%) of these Patients' records.

The mean time for CGA application was 52 min (ranging from 30 to 100 min). In our opinion, the procedure is important for the elderly and can be feasible in primary care, especially if a scheme of a first one-hour consultation followed by a 20–30 min return session is adopted. This model could improve the identification of health problems and is unlikely to overload the health care system. Options for shortening the length of the consultation could also be considered, e.g. altering the number of tests included in the screening instrument (changing GDS 15 to GDS 5, for instance); improving efficiency through pre-consultations by a nurse (not performed in the present study); or carrying out routine consultations by certain specialists (e.g. ophthalmologist).

Conclusion

In the present study, a comprehensive geriatric assessment diagnosed an average of 6 additional medical problems per patient, and was a straightforward procedure with high acceptability by patients. Therefore, the comprehensive geriatric assessment proved an important tool for GPs.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.