|Year : 2014 | Volume
| Issue : 2 | Page : 101-107
Sociodemographic factors influencing compliance of medication in an urban OPD setting
Jayanti P. Acharya, Indranil Acharya
Department of Community Medicine, Bhaskar Maedical College, Moinabad, Hyderabad, Andhra Pradesh, India
|Date of Web Publication||20-May-2014|
Jayanti P. Acharya
G-3, Kalyan Srinivasa Residency, Bank Colony, RK Puram, Secunderabad - 500 056, Andhra Pradesh
Source of Support: None, Conflict of Interest: None
The present study was conducted with an aim to assess the relationship between major sociodemographic factors and compliance to prescribed drugs, amongst outpatients in an urban health center in a big city. A cross-sectional, community-based study was conducted in the city of Ahmedabad. Random sampling was used to include 250 cases, who were interviewed in an outpatient department (OPD) set-up, using a semistructured pretested proforma. Acute upper respiratory infections (20.8%) and fever (15.2%) were the most common diagnoses. These patients were followed-up for three revisits. Twenty-eight cases (11.2%) out of the 250 were lost to the study after two follow-ups. These were mostly in the younger age groups, female, in upper social class, and literate. Another startling revelation was that, a very high majority (86-96%) reported at later dates than when they were called to the health center. Analysis of certain sociodemographic characteristics revealed that most of the respondents were in the younger age groups with a female preponderance, had had some form of formal education, and were mostly in the social middle class. Scrutiny showed that while age, education, and social class had an almost inverse relationship with compliance, sex did not influence compliance.
Keywords: Compliance, follow-up, lost cases, late reporting, sociodemographic profile
|How to cite this article:|
Acharya JP, Acharya I. Sociodemographic factors influencing compliance of medication in an urban OPD setting. J Sci Soc 2014;41:101-7
| Introduction|| |
Diseases are time immortal. They do not spare most of us during our lifetimes. Therefore, the sick seek remedial measures in different systems of medicine, to get relief from illnesses.
The medical regimen is ordinarily a set of instructions and recommendations prescribed by the doctor, is followed by a patient for alleviation of his illness. Practice of it is called compliance. 
According to the level of practice, compliance towards the prescribed regimen is described as either:
- Full or complete: When all the instructions by the doctor are carried out,
- Partial or moderate: When the regimen is followed in part, and
- Noncompliance: When very few or none of the doctor's instructions are followed. ,
Partial and noncompliance are very important public health problems since they are associated with: Wastage of resources, resistance of pathogens to antimicrobials, adverse drug reactions, prolongation of the illness, and other serious health hazards.  Thus, these lead to unsuccessful treatment; a significant problem faced by doctors today. Noncompliance may vary from 15% to a high 90% in developing countries. ,,
The yoke of illness in a developing economy like ours' is mammoth. Healthcare resources are scarce as compared to the growing population. Qualified manpower is also in short supply. In such a scenario, morbidity and mortality can definitely be reduced through timely treatment, compliance to prescribed medication and other instructions, regular follow-ups on dates allocated, and other preventive measures.
Cases lost to follow-ups (dropouts) and those reporting at a later than allocated date also adds to the noncompliance or partial compliance load. This chunk too needs to be tackled with seriousness.
The patients' sociodemographic profile and its relationship to subsequent compliance of medication is an oft-neglected and less touched upon area. Factors like social class, literacy level, and occupation have been known to affect compliance and subsequent outcomes. Importance of the involvement of families of patients in ensuring compliance to treatment must be given its due. 
Also, most of the studies on drug compliance have been carried out in developed countries ,,,, and have been mostly been done on chronic illnesses like heart diseases, , chronic suppurative otitis media,  and hypertension. , Fewer studies have been conducted in India. ,, Studies on acute and/or common illnesses seen in an outpatient department (OPD) setting are rare.
Outcomes of studies relating the patient profile to compliance especially for acute illnesses; could help identify factors to improve compliance; thus ensuring faster cures, lesser loss in man-days, and a reduced morbidity load in the community.
| Materials and methods|| |
The city of Ahmedabad lies at 23.02°N latitude and 72.37°E longitude, in India. Its cantonment area is located in the northeastern part, spread over 5.6 km 2 area. The residential areas are within the cantonment as well as in adjacent localities of Sadar Bazaar, Sardar Nagar, Meghani, Nagar, etc.
The study was carried out over a period of 3 months in areas around the cantonment, in one of the health centers having basic facilities like a laboratory, X-ray, and a minor operation theater (OT). The average daily patient load of this center varied from 40-50 cases a day. A medical officer, assisted by paramedical staff was available during the study. The medicines were dispensed free through the health center's pharmacy.
Random sampling method was utilized and a total of 250 cases that required three follow-ups, were selected for the present study. In case of children below 15 years age, the person accompanying the child was interviewed. Those 15 and above answered the questions themselves.
A semistructured pretested proforma was utilized. Oral questions were asked and the proforma filled-up by the authors themselves. A leading question as to whether the patient had consumed all or some or none of the drugs prescribed was asked to categorize the compliance category. For this the authors had to depend entirely on the patients' or guardians' integrity. Information on sociodemographic profile of the patients, the various diagnoses, and subsequently their compliance status and its relationship with the sociodemographic profile was recorded.
The data were complied, presented, and statistically analyzed.
| Observations/Results|| |
Out of a total 3,395 new OPD cases in the study period, 250 (3.91%) who required minimum three follow-ups, were included in the present study based on a random sampling technique. The first visit was termed the I follow-up; the 2 nd , II follow-up; and the last was called as III follow-up.
Although various forms of illness were observed; the ten leading illnesses diagnosed in the first visit or the I follow-up, were [Table 1]:
All the patients in the present study were instructed to report at a specific date after their first (I follow-up) and then second visits (II follow-up). [Table 2] gives the distribution of cases according to follow-up coverage.
It was observed that at first and second follow-ups all the cases reported either on time or after some delay. They were followed-up and given a subsequent date for the third follow-up. However, at the third follow-up, 28 (11.2%) cases were considered "lost to the study", after they did not report even after a week from the given date.
Late reporting, if any, at each follow-up was assessed and results are summarized in [Table 3]. Analysis revealed that a majority reported late as compared to few who reported on the date they were asked to report. The late reporting according to follow-up status varied between 86.0 (215 cases) and 96.0 (213 cases). The reporting on due date varied from 4.0 (nine cases) to 14.0% (35 cases). Reporting thus showed an inverse relationship with follow-up status. The late reporting in any of the follow-ups, varied between 1 and 7 days. The average duration of late reporting cases varied between 2.577 ± 1.149 and 3.068 ± 1.315 days.
|Table 3: Distribution of cases according to their reporting status at each follow-up|
Click here to view
Sociodemographic profile and compliance
Compliance was categorized as full, partial, or non-compliance according to the patients' own submissions on asking the leading question as to whether they consumed all the drugs prescribed, consumed only some of the drugs, or used none of them, respectively. Bias due to patients' own perceptions had to be overlooked. Data on four key sociodemographic aspects: Age, sex, education, and social class was collected and analyzed in relation to patients' compliance.
The age distribution reveled that more (163, 65.2%) cases were under 15 years of age group, while only 87 (34.8%) were 15 years or above [see [Table 4] below]. Those below 15, were in most cases accompanied by a parent or some elder relative or neighbor.
Full compliance was observed to be more in the less than 15 years age group (59.4-65.03% for follow-ups and 62.3% in weighted) than in more than 15 years age group (45.6-68.9% for follow-ups and 55.3% for weighted). Partial compliance on the contrary, was more in more than 15 years age group (31.1-54.4% for follow-ups and 43.5% for weighted) as compared to the less than 15 years age group (34.97-40.6% for follow-up and 37.5% for weighted). Similar findings were seen for the four noncompliant cases, there was one case in the less than 15 years age-group; whereas, three cases were there among those 15 or above. Of the 28 lost to follow-up, 20 were from less than 15 years age group, and rest eight from those 15 years or above. [Table 5] below gives the relationship between compliance status and age-groups.
The age distribution according to sex showed a similar tendency. In the study, there were 104 (41.6%) males as compared to 146 (58.4%) females. The sex ratio of cases was thus almost 1,400 females per 1,000 males. When studied according to age, the sex ratio of cases under 15 years age group, was about 1,200 females per 1,000 males and for 15 years and above age group it was 1,900 females per 1,000 males. Detailed information on age and sex distribution in shown in [Table 4], below.
Full compliance was marginally more amongst males (54-67.3%) as compared to females (53.6-65.8%). Partial compliance was marginally more in the females (34.2-46.4%) as compared to males (32.7-44.3%). Seven males and 19 females were lost to follow-up. Among the noncompliant, one was male and rest three were of the female sex. [Table 6] gives further details.
For cases in age group less than 15 years, the education of the father was considered and noted, while for age group of 15 years and above, the education status of the individual was considered. [Table 7] underneath gives the details.
|Table 7: Distribution of cases according to educational level of fathers in under 15-year-old cases and of patients in 15 years and above age group|
Click here to view
The education status of fathers in cases of age group of less than 15 years (n = 163), revealed that 13 (7.9%) were illiterate; seven (4.2%) were just literate; 27 (16.6%) had had primary education, 50 (60.7%) secondary, 43 (26.4%) higher secondary education; and 18 (11%) were graduates and 5 (3.2%) were postgraduates.
The literacy status of cases in the age group 15 years and more (n = 87), revealed that 10 (11.5%) were illiterate, seven (8%) were just literate, and that 28 (32.2%) had primary, 22 (25.3%) secondary, 16 (18.4%) higher secondary education, two (2.3%) were graduates, and two (2.3%) postgraduates. The overall educational profile was in favor of educated group (213 or 85.2%), which in descending order was followed by the illiterate (23 or 9.2%) and the just literate (14 or 5.6%) groups.
As regards compliance to medication, full compliance was highest among the illiterates (65.2-91.3% for follow-up and 75.8% for weighted). This was followed by the just literate group where full compliance ranged from 57.11 to 85.7% for follow-ups and 71.8% for weighted, and lastly by the educated group where it ranged from 51.8 to 62.4% for follow-ups and 57.4% for weighted.
Partial compliance showed a reverse trend and was maximum in the educated group (37.6-48.2% for follow-ups and 41.9% for weighted) followed by the just literate group (14.3-42.9% for follow-ups and 28.2% for weighted), and lastly by the illiterate group (8.7-34.8% for follow-ups and 24.2% for weighted). The four (1.9%) noncompliant cases seen in the follow-ups were amongst the educated. [Table 8] summarizes all this data.
Modified Prasad's Socioeconomic Classification  was followed and suitably modified as per today's income strata, while denoting social class of the cases, in the present study. Majority of them (215.86%) were from social classes II (44.4%) and III (41.6%). Remaining 35 (14%) cases were from social classes I (11.2%) and IV (2.8%). The population studied therefore represents mostly the middle income group/social class. [Table 9] here elaborates the data.
The weighted full compliance was maximum in social class II (63.0%). In a descending order, decrease in compliance was observed in social class II (60.4%); then social class IV (57.1%), and finally social class I (48.75%). Alternatively, the weighted partial and noncompliance were maximum in social class I (51.25%); and the descending order decrease in compliance was next observed in social class IV (42.9%), social class II (40.6%), and social class III (37.0%). The observations are tabulated below in [Table 10].
|Table 10: Compliance status of cases in relation to social classifi cation|
Click here to view
| Discussion|| |
As mentioned, compliance to medication is a relatively new area of study and not much has been done in this field, especially in a developing country like India.
In the present study, 250 patients were followed-up from first through third visits. The common profile of the case/attender was of a young and educated person, belonging to the middle social class. There were more female participants than males, in this study.
Loss of cases during follow-up in the present study was 11.2%, all at the III follow-up, almost similar to reports by Linn et al.  and Hungerbuhler et al.  More cases were 'lost to follow-up' from the younger age group of less than 15 years (12.3%), more females (14.4%) were lost to the study than males, more lost cases belonged to the social class I (14.3%) and to the just literate group (21.04%). Reasons for same need to be studied in detail and discrepancies in part of the medical authorities as well lack of awareness if any, need to be looked into. Studies that have analyzed 'lost cases' in a similar manner, could not be traced.
Late reporting of cases was found to be rampant. It varied from 86 to 96% and ranged from 1 to 7 days. The reasons for these need to be analyzed in further studies of similar nature and the importance of complying with the doctors' orders by reporting on due date for a follow-up needs to be stressed upon time and again on the clientele.
Thus, two very important aspects revealed by the study and areas of real concern were losing a patient to follow-up and reporting later than given date for a follow-up. On one hand, while it is always stressed that health facilities need to be more patient-friendly, the aspect of attitudinal change to access such facilities on time and every time needs to be emphasized upon to the patients and their families. The young, the female sex, and those in social class I perhaps without proper knowledge on the prescribed drugs or due to a slapdash approach, must be able to appreciate these factors. Then only, improvement in compliance levels would follow and cure rates augmented.
Though Linn et al.,  reported older patients to be more complaint than younger ones, the present study revealed slightly more (62.3%) compliance in patients in the less than 15 years age group than in the patients of 15 years and above (55.3%). This was perhaps for the reason that defense services' ladies/personnel was more conscious of the importance of medication and did administer the prescribed drugs to their children as per schedule. Noncompliance was observed to be more in the above 15 age group, perhaps due to the notion that prolonged medication would give rise to some after or side-effects; while some in this group were partially compliant as relief from symptoms possibly prompted an abrupt stoppage of medication.
Proportionate distribution of cases according to full, partial, and noncompliance was observed to be nearly similar in either sex, pointing to the fact that sex might hardly have any influence on compliance to medication. In poorer and/or conventional communities, the compliance levels drop in females due to them forgetting to take the drugs due to household work and their family's negligence (Jain et al.,  ).
Educational level was expectedly found to have an inverse relationship with compliance in the present research. The illiterate and just literate people were much more fully compliant (74.2%) as compared to the educated (57.4%). Similarly, partial and noncompliance was more in the latter, that is, educated group (43.8%) as compared to the former two (25.8%). Davis  has documented in his study on cardiac patients, that higher education led to failure in compliance; whereas, Johnson  found just the opposite in his study of heart patients. Conversely, Francis et al.,  found no significant relation between education and compliance, when college educated mothers had 43.6% compliance compared to an almost similar 38.5% amongst others who were less educated in his study.
Partial and noncompliance was again predictably highest amongst upper social class patients (51.4%) and almost similar among the patients belonging to the other social class (37.0-42.9%). Francis et al.,  found no significant relationship between social class and compliance in his study of 800 patient-OPD visits, though it was reduced for class IV and V (31.5%) as compared to I and II (44.8%).
As per existing literature, compliance to prescribed medication is not much influenced by certain demographic features like the family type, family size, and martial and residential status; though Venugopal et al.,  do relate compliance to occupation. Hence, these factors were not analyzed in the present study. Gordis et al., Davis, and Francis et al., etc., ,, have also emphasized upon the fact that these sociodemographic factors were unrelated to compliance. Mattar et al.,  and Hulka et al.,  have also found no relationship between race, marital status, occupation, etc., and compliance. Kasl and Cobb  had also stated that most compliance studies had focused too much on demographic variables rather than attitude and perceptions.
Still a sincere attempt was made in the present study to relate certain sociodemographic factors to compliance and the outcome observations have been convincing. Age, literacy, and social class did influence compliance to a noteworthy extent. Since the study was conducted among a disciplined and somewhat informed sample of population, the differences in complete and incomplete compliance were not wide ranging. More studies in different communities and comparisons between communities divergent in these sociodemographic variables would certainly expose loopholes which need to be taken cognizance of and timely preventive measures put in place to improve compliance to prescriptions as well as the physician's instructions. Reasons for partial and noncompliance among the educated upper social classes is a matter of concern and must be analyzed energetically. Issues like early disappearance of symptoms, a feeling of betterment, a perspicacity about side-effects, and less faith in the treating doctor could lead to partial cure and/or frequent relapses must be researched into.
| References|| |
|1.||Mazzulo JM, Lasagna L, Griner PF. Variation in interpretation of prescription. Instructions JAMA 1974;227:929-31. |
|2.||Francis V, Korsh BM, Morris MJ. Gaps in doctor-patient communication. N Engl J Med 1969;280:535-40. |
|3.||Latiolas CJ, BerryCC. Drug intelligence and clinical pharmacy. 1969;3:279. |
|4.||Kiyingi KS, Lauwo JA. Drugs in the home: Danger and waste. World Health Forum 1993;14:381-3. |
|5.||Linn MW, Linn BS, Stein SR. Satisfaction with ambulatory care and compliance in older patients. Med Care 1982;20:606-14. |
|6.||Davis MS. Variation in patients′s compliance with Doctor′s Advice: An Empirical Analysis of patterns of Communication. Am J Public Health Nations Health 1968;58:274-88. |
|7.||Venugopal R, Narasimha Murthy NS, Gopinath D. Factors influencing compliance to treatment among people with chronic illness in an urban area of South India. Int J Biol Med Res 2012;3:1495-7. |
|8.||Williams A, Manias E, Walker R. Interventions to improve medication adherence in people with multiple chronic conditions: A systematic review. J Adv Nurs 2008;63:132-43. |
|9.||Piette JD, Heisler M, Wagner TH. Cost-related medication under use among chronically Ill patients: The Treatments people forgo, how often, and who is at risk. Am J Public Health 2004;94:1782-7. |
|10.||Mattar ME, Marketta J, Yajje SJ. Pharmaceutic factors affecting paediatric compliance. Paediatrics 1975;55:101-8. |
|11.||Becker MH, Drachman RH, Kirscht JP. Prediciting mother′s compliance with paediatric medical regimen. J Paediatrics 1972;81:843-54. |
|12.||Elling R, Whittemore R, Green M. Patient participation in a paediatric program. J Health Human Behav 1960;1:183. |
|13.||Munger M, Van Tussel BW, LaFleur J. Medication non-adherence: An unrecognised cardio-vascular risk factor. MedGenMed 2007;9:58. |
|14.||Hungerbuhler P, Bovet P, Shamlaye C, Burnand B, Waeber B. Compliance with medication among out patients with uncontrolled hypertension in the Seychelles. Bull World Health Organ 1995;73:437-42. |
|15.||Morgado M, Rolo S, Macedo AF, Pereira L, Castelo-Branco M. Predictors of uncontrolled hypertension and anti-hypertensive medication non-adherence. J Cardiovasc Dis Res 2010;1:196-202. |
|16.||Jain L, Saxena S, Sen S. The fate of yur prescription. Indian Pediatr 1985;22:367-70. |
|17.||Arya S, Naidu AN. Community awarness attitudes and utlilisation of primary health centre in rural Hyderabad. Indian J Commu Med 1983;8:13-21. |
|18.||Prasad BG. Changes proposed in social classification of Indian families. J Indian Med Assoc 1970;55:198-9. |
|19.||Johnson WL. Conformity to medical recommendations in coronary heart disease. American Sociological Association, Chicago, Sep 2, 1965. |
|20.||Gordis I, Markowitz M, Libenfield AM. The inaccuracy of using interviews to estimate patient reliability in taking medications at home. Med Care 1969;7:49-54. |
|21.||Hulka BS, Casual JC, Kupper LL, Burdette JA. Communication, compliance and concordance between physicians and patients with prescribed medications. Am J Public Health 1976;66:847-53. |
|22.||Kasl SV, Cobb S. Health behaviour, illness behavior and sick role behaviour. Arch Environ Health 1966;12:246. |
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8], [Table 9], [Table 10]