Social environment, morbidity and use of health care among people with diabetes mellitus in Trinidad

BACKGROUND: This study aimed to identify social characteristics associated with higher levels of morbidity from diabetes and their relationship to health care utilization. METHODS: During a 6-month period 1149/1447 (79 percent) subjects admitted to Port of Spain Hospital, Trinidad with diabetes responded to a structured interview. Data collection included social factors, diabetes-related morbidity and health care utilization. Analyses were adjusted for age, sex, ethnic group and self-reported diabetes duration. RESULTS: Of 12 indicators of morbidity, nine were more frequently in subjects with no schooling compared with those with secondary education. At ages 15-59 years, nine morbidity indicators were less frequently among subjects in full-time jobs compared with those not in employment. The association of educational attainment was explained by confounding with age, sex, ethnic group and diabetes duration but five morbidity indicators were associated with employment status after adjusting for confounding. The type of water supply in the home was generally not associated with morbidity. Each of the indicators of lower socioeconomic status was associated with less use of private doctors and with more use of government health centres. CONCLUSIONS: Morbidity from diabetes was greater in groups with lower socioeconomic status. While morbidity associated with lower educational attainment was mostly explained by older age; the results suggested the possibility that diabetes may contribute to unemployment of those in the labour force. Private care was less accessible to social groups with higher levels of morbidity and the availabiltiy of government funded health services was important for reducing inequalities in health care utilization.(AU)

Diabetes care in middle-income countries: a Caribbean case study

Many middle-income countries now have a high prevalence of diabetes and need to address the problem of providing care for people with diabetes within limited resources. This study evaluated standards of preventive care in primary settings in three Caribbean countries. We studied case records at 17 clinics in 15 government health centres and 17 private general practitioners' offices in Barbados Trinidad and Tobago and Tortola (British Virgin Islands). A census of all attenders over 4 to 7 week period identified 1661 attenders with diabetes mellitus, approximately two-thirds were women with a median age of over 60 years. Overall 676/1342 (50 percent) had 'poor' blood glucose control (> or + 8 mmol 1-1 fasting or > or = 10 mmol 1-1 random). The proportion with BP > or = 160/95 mmHg or receiving treatment for hypertension was 943/1661 (57 percent), of whom 781/943 (83 percent) were prescribed drug treatment. Among those treated for hypertension only 181/781 (23 percent) had blood pressures , 140/90 mmHg. Surveillance for complications affecting the feet (11 percent) or eyes (2 percent) was not performed systematically in any setting. Only 533 (32 percent) had recorded dietary advice and 79 95 percent) had recorded exercise advice in the last 12 months. To begin to address some of these problems at the regional level, we incorporated results from this survey into a series of workshops held in collaboration with health ministries in 10 Caribbean countries, with participants from 13 countries. At these workshops health care workers participated in the process of developing guidelines for diabetes management in primary care. The guidelines have subsequently been widely disseminated through health ministries and non-governmental organizations in the region. Further research is needed to evaluate the effectiveness of this approach, the constraints of diabetes care, and the most cost-effective means of addressing them (Au).

Counting the cost of diabetic hospital admissions from a multi-ethnic population in Trinidad

Many middle-income countries are experiencing an increase in diabetes mellitus but patterns of morbidity and resource use from diabetes in developing countries have not been well described. We evaluated hospital admission with diabetes among different ethnic groups in Trinidad. We compiled a register of all patients with diabetes admitted to adult medical, general surgical, and ophthalmology wards at Port of Spain Hospital, Trinidad. During 26 weeks, 1447 patients with diabetes had 1722 admissions. Annual admission rates, standardized to the world population, for the catchment population aged 30-64 years were 1031 (95 percent CI 928 to 1134) per 100,000 in men and 1354 (1240 to 1468) per 100,000 in women. Compared with the total population, admission rates were 33 percent higher in the Indian origin population and 47 percent lower in those of mixed ethnicity. The age-standardized rate of amputation with diabetes in the general population aged 30-60 years was 54 (37 to 71) per 100,000. The hospital admission fatality rate was 8.9 percent (95 percent CI 7.6 percent to 10.2 percent). Mortality was associated with increasing age, admission with hyperglycaemia, elevated serum creatinine, cardiac failure or stroke and with lower-limb amputation during admission. Diabetes accounted for 13.6 percent of hospital admissions and 23 percent of hospital bed occupancy. Admissions associated with disorders of blood glucose control or foot problems accounted for 52 percent of diabetic hospital bed occupancy. The annual cost of admissions with diabetes was conservatively estimated at TT10.66 million (UK 1.24 million pounds). In this community diabetes admission rates were high and varied according to the prevalence of diabetes. Admissions, fatalities and resource use were associated with acute and chronic complications of diabetes. Investing in better quality preventive clinical care for diabetes might provide an economically advantageous policy for countries like Trinidad and Tobago. (AU)

Counting the cost of diabetic hospital admissions in North Trinidad - abstract

A register was compiled of all patients with diabetes mellitus who were admitted to medical, general surgical or opthalmology wards at the Port-of-Spain General Hospital over a 26-week period form October, 1993 to April, 1994. Admission rates declined with increasing distance from the hospital, but increased with age, were higher in women than in men, and higher in Indo-Trinidadians than in Afro-Trinidadians. There were 1722 diabetic admissions of whom 1334 (77 percent) were medical, 302 (18 percent) surgical and 86 (5 percent) opthalmology. Frequent diagnoses on admission were: hyperglycaemia, 19 percent; hypoglycaemia, 12 percent; stroke and myocardial infarction, 8 percent each. The median length of stay was 4 days (interquartile range, 2 to 9 days). Diabetic inpatients occupied an estimated 26,659 hospital bed days per year, with an estimated hospital cost > TT$10.6 million (US$1.8 million). Patients admitted with disorders of glycaemic control or foot problems occupied an estimated 13,913 bed days per year with an approximate annual cost of TT$5.5 million (US$0.96 million). The mean cost of one diabetic admission was approximately TT$3,096 (US$538). This sum would cover the cost of one year's treatment in a government primary care setting for up to 9 persons with diabetes mellitus. Many admissions might be avoided with improved preventive management in primary health care settings. Improving the quality of primary care for diabetes mellitus might free considerable hospital resources for alternative use (AU)

Social inequalities in health status and care practices among patients admitted to hospital with diabetes mellitus in Trinidad - abstract

Of 1149/1447 (79 percent) patients admitted to the General Hospital, Port-of-Spain, with diabetes mellitus in a 6-month period, 138 (12 percent) had no schooling (NS), 782 (68 percent) had only primary education (PE), 119 (17 percent) had secondary education or higher (SE), 30 (3 percent) were not classified. Compared with patients with SE, those with NS were older, were more likely to be women, or of East Indian descent and to have effort chest pain, a history of `stroke' or `heart attack' (Odds Ratio (OR) 3.51 (95 percent CI 1.97 to 6.24)), neuropathic symptoms (OR 1.72 (1.11 to 2.68)) and vision defects (OR 2.97 (1.28 to 6.87)). Variations in the prevalence of these conditions were mostly explained by confounding with increasing age, female sex and East Indian ethnicity. Patients with NS were more likely to attend a health centre than a private doctor as their usual source of care (OR 4.51 (2.76 to 7.35)), were less likely to have attended a dietitian (OR 0.53 (0.34 to 0.83)), and were less likely to test for blood glucose at home (OR 0.14 (0.03 to 0.61)). Adjusting for age, gender and ethnic group did not explain social inequalities in the usual source of care, attendance at a dietitian or use of home blood glucose monitoring. Our data are consistent with the `inverse care law': those most affected by diabetic complications had more limited access to health care (AU)

Drug prescribing for diabetes mellitus in primary care: a comparison of public and private practice in three Caribbean countries - abstract

Examining the appropriateness of drug prescribing for people with diabetes mellitus in Caribbean countries is important because of the high cost of drug treatment, the potential for improved control of the disease and the possibility of reducing adverse side effects of treatment. This study examined patterns of drug prescribing for diabetes mellitus in public and private sector primary care settings in three Caribbean countries. The sample included 690 patients in Barbados (BDS, 24 percent private), 791 in Trinidad and Tobago (TT, 13 percent private) and 180 in Tortola (BVI, 31 percent private). Patients treated in public health care facilities were prescribed significantly more drugs than those treated in private practice. Few patients had diabetes mellitus managed by diet alone (8 percent public, 15 percent private). Metformin was rarely used as single agent therapy (3 percent public, 6 percent private). Most patients were treated with sulfonylurea drugs alone or in combination with metformin (75 percent public, 67 percent). The proportion of sulfonylurea prescriptions for chlorpropamide varied (Public: BVI 80 percent, TT 60 percent, BDS 10 percent; Private: BVI 41 percent, TT 28 percent, BDS 7 percent) as did prescriptions for gliclazide and glipizide (Public: BDS 41 percent, BVI 3 percent, TT 1 percent; Private: BVI 51 percent, BDS 46 percent, TT 19 percent). A high proportion of patients were treated for hypertension (public 49 percent, private 40 percent). In private practice, ACE inhibitors and diuretics were the most frequently prescribed drugs. In the public sector, Brinderin accounted for 53 percent of prescriptions in TT while thiazides, methyldopa, betablockers and ACE inhibitors were the most frequently prescribed drugs in BDS and BVI. These variations in prescribing practice among countries of the region suggest that factors other than patients' needs or the cost effectiveness of treatment are important in determining prescribing practices. Individual countries should examine how efficiency and effectiveness of drug use could be improved (AU)

Quality of diabetes care in Trinidad and Tobago: surveillance for risk factors and complications - abstract

Long-term complications of diabetes mellitus cause increasing disability and health care costs in many Caribbean countries. Clinical care of people with diabetes mellitus should detect and treat arterial hypertension and cigarette smoking which are risk factors for the development of long-term complications. Patients should also be screened for signs of complications of diabetes mellitus so that these may be treated at the earliest stage. This study aimed to evaluate the extent to which these objectives were achieved by health care services in Trinidad and Tobago. Data were abstracted from the case notes of 1723 patients with diabetes mellitus attending two hospital clinics, nine government health centres and eight general practitioners. Blood pressure (BP) was measured within the preceding 12 months for 94 percent of patients but 30 percent had BP o160/95 mm Hg despite anti-hypertensive drug use in 41 percent of all patients. Among diabetic patients treated for hypertension, only 40 percent had BP ó140/90 mm Hg. Smoking habits were recorded for 67 percent of hospital patients but only for 15 percent of primary care patients. For cases with available date, 17 percent were current smokers and 8 percent were ex-smokers. Examination of the feet was recorded for 14 percent of patients within the last twelve months and fundoscopy was recorded in only 4 percent. The BP was regularly measured and a high proportion of patients were treated for hypertension but BP control was unsatisfactory. Cigarette smoking was a significant and neglected problem. Screening for signs of diabetic complications was not performed systematically. Improved strategies for the preventive care of people with diabetes mellitus require implementation (AU)

Intermediate outcomes of primary care for diabetes mellitus: a comparison of public and private practice in three Caribbean countries - abstract

This study examined the quality of monitoring and control of obesity, hyperglycaemia and hypertension among 1661 attenders with diabetes mellitus at public health centres and private general practitioners in Trinidad and Tobago (791 (13 percent private)), Barbados (690 (24 percent private)) and Tortola (180 (31 percent private)). Weight was recorded in the preceding 12 months for 83 percent but heights were recorded only for 2 percent. Blood glucose was measured in the last 12 months for 93 percent in Tortola, 82 percent in Barbados and 36 percent in Trinidad. Blood pressure was measured in the preceding 12 months for 95 percent of patients. Compared with Barbados the proportion overweight (>72 kg female and >85 kg male) was higher for clinic attenders in Tortola (OR 1.44 (95 percent CI 1.02 - 2.04)) and less in Trinidad and Tobago (0.60 (0.48 - 0.76). The proportion with 'poor' blood glucose control (8 mmol/l fasting or 10 mmol/l random) was higher at clinics in Trinidad (1.61 (1.27 - 2.04) than in Barbados and lower in private than in public practice (10.57 (0.44 - 0.75)). The proportion with untreated hypertension (BP 160/95 mm Hg) was less for clinics in Tortola than in Barbados (0.13 (0.04 - 0.41) as was the proportion with treated but uncontrolled (BP > 140/90 mm Hg) hypertension (0.29 (0.18 - 0.47). Private practice attenders were more likely to have untreated hypertension (2.07 (1.42 - 3.03) or treated but uncontrolled hypertension (1.87 (1.23 - 2.83)) when compared to public practice. These variations in intermediate outcomes of diabetes care might be partly explained by differences in case mix but emphasise as priorities the control of obesity for clinics in Tortola, the control of hyperglycaemia for clinics in Trinidad and Tobago, and the control of hypertension in private GPs' offices when compared with health centres (AU)