Assessing the Quality of Life of Filipino Cancer Patients: A Survey of Filipino Oncologists

Study Design and Setting

After obtaining approval from the University of the Philippines Manila Ethics Review Board, a cross-sectional survey among clinical oncologists and oncologists-in-training, practicing, or training in the Philippines was performed. The HRQoL assessment form, a quantitative and qualitative questionnaire, was used to obtain data, including socio-demographic information, questions on how HRQoL is assessed, the oncologists’ general attitude toward the concept of HRQoL, and the different factors that affect HRQoL assessment in clinical practice. Respondents in the study were recruited through email channeled through the specialty societies (e.g., Philippine Society of Medical Oncology, Philippine Society of Radiation Oncology, Philippine Society of Gynecologic Oncology, Philippine Society of Hematology, Philippine Society of Surgical Oncology) after being granted permission from their Head of Office/President. Purposive sampling during specialty conferences or gatherings was also done.

Questionnaire Development

The questionnaire used was adapted from previous studies obtained from review of literature.^{12,25,26,27,28,29,30,31} The instrument consisted of four parts: (1) study introduction and objectives to fully inform the participant what the study is all about; (2) demographic data including age, gender, oncologic specialty, level of training, and working environment; (3) ten closed questions about their attitude toward HRQoL (Likert-scales); (4) seven questions regarding their assessment of HRQoL in clinical practice.

Content validity of the questionnaire was evaluated by a panel consisting of five oncology consultants, and five oncologists-in-training. Modifications were made according to their recommendations before subjecting to preliminary testing.^32,33,34

Face validity and reliability were assessed by preliminary testing on a convenience sample of 30 oncologists/oncologists-in-training in the study setting who were excluded from the main study. Respondents were asked for comments on the readability of the items, and for suggestions to make the questions easier to understand, if needed. Participants were asked to accomplish the questionnaire again after 2 weeks to obtain test–retest stability. The survey questions were deemed homogenous, with acceptable internal consistency (Pearson’s correlation test-retest coefficient 0.9, and Cronbach’s α reliability coefficient 0.85).

Construct validity, the degree to which inferences regarding a theoretical construct can be made based on certain operationalizations, was evaluated using a two-stage sorting procedure consisting of an unstructured followed by a structured sorting exercise. This was used to evaluate for both convergent and discriminant validity, which is important because several items on the questionnaire were self-developed based on the current study objectives. The hit rate was 90% after the second stage of the sorting procedure and was deemed acceptable.³⁵ Validity was further determined through factor analysis, with a minimum factor loading of 0.5 for each item required for all items in the final survey questionnaire. Results from the validity and reliability testing were used to determine the final list of questions that were included in the questionnaire.

The final survey consisted of the following three parts:

• Demographic data (age, gender, field of oncology, level of training, working environment, bulk of cancer type seen in practice);

• Ten questions about the Filipino oncologists’ attitude toward HRQoL assessment;

• Seven questions regarding the Filipino oncologists’ assessment of HRQoL in clinical practice.

Population Selection

Clinical and Demographic Characteristics

Clinical data analyzed included age, gender, field of oncology (medical, surgical, radiologic, gynecologic, hematologic), level of training (consultant or fellow-in-training), working environment (government, private, or both), and bulk of cancer type seen in clinical practice.

Statistical Analysis

Demographic characteristics were summarized using descriptive statistics (mean and standard deviation). Attitudes and perceptions on HRQoL assessment were described using frequencies and percentages.

Logistic regression analysis was used to determine the statistically significant predictors of use of various means of assessment of QoL. Backward elimination method was the variable selection procedure used in this analysis. A probability to remove of 5% was used as cutoff in determining variables to be retained in the final model. Both descriptive and inferential statistics were calculated using Stata 14 SE.

Clinicodemographic Profile

A total of 312 Filipino oncologists, from five subspecialities, participated in the study (Table 1). The mean age of the oncologists was 38 years old (standard deviation of 7.73 years). Majority were female (63%), practiced in Luzon (80%), practicing oncology for an average of 6 years.

Attitude of Filipino Oncologists on HRQoL Assessment

Ninety-six percent of the participants reported that HRQoL is important for clinical work (Table 2). Only 42% of the oncologists disagree that HRQoL is a vague term and about two-thirds perceived HRQoL as suitable for daily clinical use. Almost all oncologists (97%) perceived that HRQoL assessment is valuable in patient management. While majority (91%) of the oncologists perceive that validated HRQoL instruments are useful for QoL assessments, more than two-thirds (69%) of the oncologists agree that verbal HRQoL assessments are generally enough. Furthermore, 39% of the oncologists reported that their patients do not accept HRQoL questionnaires and another 39% of the oncologists were unsure if their patients accept HRQoL questionnaires.

More than three-fourths (78%) of the oncologists do not know which HRQoL assessment tool to use for Filipino patients, and 78% of the oncologists perceived that HRQoL questionnaires are too extensive and too lengthy for routine. Lastly, more than half (54%) of the oncologists reported that they are not sufficiently trained to assess HRQoL.

Implementation of HRQoL Assessment by Filipino Oncologists in the Clinical Setting

The oncologists reported that the most common motivations for assessment of their patients’ QoL were to support a therapy choice (86%) and for baseline assessment (81%) (Table 3).

The most common means of assessment of QoL employed by the participants was interview using nonstandardized queries (Table 4).

The most commonly reported validated HRQoL questionnaires that oncologists are familiar with and comfortable to use were the Eastern Cooperative Oncology Group (ECOG) score (76%), and the Karnofsky Index (68%) (Table 5).

Majority of the oncologists (89%) reported that they will be more confident when local HRQoL assessment guidelines are present, and 93% agreed to the use of a unified HRQoL assessment tool for Filipino cancer patients (Table 6).

More than half (58%) of the oncologists utilized HRQoL assessment for advanced/metastatic cancers, while 39% used it for both early and advanced stage disease (Table 7). Most (85%) oncologists preferred that the optimal number of questions for a HRQoL assessment tool should be between 5 and 15 questions (Table 8).

Factors Influencing the Filipino oncologists’ HRQoL Assessment

The significant predictors of HRQoL assessment among oncologists were field of oncology and year in practice/training (p < 0.05, Table 9). In particular among the fields of oncology, the odds of doing HRQoL assessment were highest among radiation oncologists (6.6 times higher compared with medical oncologists). Meanwhile, the odds of doing HRQoL assessment decreased by 16% per year increase in the oncologists’ years in practice/training.

Attitude of Filipino Oncologists on HRQoL Assessment

The general positive interest and attitude of Filipino oncologists in assessing HRQoL of their patients are similar to those observed in physicians of other countries.^36,37,38,39

The results of our study also indicated that a recognition of the utility and importance of HRQoL assessment did not necessarily translate to routine use in clinical practice. It is striking to note that there was an overwhelming agreement that HRQoL assessment is important, with two-thirds of the participants indicating its suitability for daily use. However, the participants have been referring to verbal assessments of HRQoL instead of using standardized HRQoL instruments.

From the physician’s perspective, it is more convenient to ask random questions instead of having to answer lengthy questionnaires that would take too much time. The hesitation to do HRQoL assessment routinely may also be due to the oncologists’ perceived notion that the instruments are too extensive and lengthy for routine clinical use.⁴⁰ Interestingly, this perception is not unique to Filipino physicians.^41,42,43 Our results also showed that approximately 80% of the oncologists were either sure that answering HRQoL was unacceptable to their patients or that they were uncertain if it was acceptable at all. The inconvenience that this would cause their patients may have driven the physicians’ hesitation in using standardized questionnaires.

It is noteworthy that, aside from the impracticability of using lengthy HRQoL assessment tools in routine clinical practice, several factors exist as to why these tools are not more frequently used. These include uncertainty as to what HRQoL tool to use, as well as a lack of training or orientation in assessing HRQoL of patients. This stresses the importance of continuing medical education, with lectures and symposia dedicated to informing and encouraging oncologists to include HRQoL assessments in clinical practice. Oncologists-in-training may also benefit from the inclusion of HRQoL assessments in their curriculum. Certainly, it is not just Filipino oncologists who feel this way; a study among Dutch physicians showed that they felt they were not sufficiently trained to using HRQoL tools either.³⁶

Implementation of HRQoL Assessment by Filipino Oncologists in the Clinical Setting

HRQoL assessment was used by the majority to support a therapeutic choice, for baseline assessment, and as part of evaluation during follow-up. This result is similar to a study among urologists in Germany,²⁹ while interestingly a study done among physicians in Italy mainly obtained HRQoL assessment for research purposes.⁴⁴

Majority assessed the patients’ HRQoL verbally using nonstandardized queries. Unfortunately, several studies have already proven that the use of standardized validated instruments for HRQoL assessment is more accurate and reliable in the management of cancer patients.^{45,46,47,48,49,50,51}

Furthermore, majority of the Filipino oncologists were familiar and comfortable to use the ECOG score and Karnofsky Index as a method of HRQoL assessment. However, studies have shown that more accurate and wholistic tools to use that would cover assessment of the cancer patients’ biological/physiological variables, symptom status, and functional status are the WHO-QOL and the EORTC QLQ-C30 questionnaires.^52,53,54,55 In fact, a study done in London suggested that HRQoL information can improve patient care by improving communication between the patient and the physician, prioritizing problems for treatment, screening for unmet needs that may warrant referral to other subspecialties, identifying patient preferences among the goals of treatment that could improve adherence, and monitoring adverse effects of treatment that may not be evident clinically.⁵⁶

Also, only 5% of the Filipino oncologists know that there is already an existing standardized and validated HRQoL questionnaire specifically for Filipino cancer patients—the UP-DOH QoL assessment tool crafted by Ngelangel in 2008.²

Majority of the Filipino oncologists agreed to the use of a unified HRQoL assessment tool for Filipino cancer patients and that they would be more confident if local guidelines for its use are present. Moreover, majority reported that the optimal number of questions in the assessment tool should be between 5 and 15 for it to be practical for daily use. These results are consistent with the findings in several studies that indicate the importance of national guidelines in the use of HRQoL assessment tools especially in encouraging physicians to incorporate it in their daily practice.^{57,58,59,60,61}

Also, one of the lessons learned during the creation of the EORTC QOL questionnaire used universally in European countries is that the length of the questionnaire should not only be complete enough to cover the different aspects of the patients’ QoL, but of equal importance is that its length and administration time should also be well accepted by the physicians for its use to be successful.^62,63,64

Factors Influencing the Filipino Oncologists’ HRQoL Assessment

Factors noted to significantly influence doing HRQoL assessment among Filipino oncologists were field of oncology and years in practice/training.

As the seniority of the oncologists increases whether in practice or in fellowship training, their use decreases by 16% per year. This might be explained by a finding in one study done in Canada which showed that 69% of the respondents (mostly >50 years old) would be more likely to base their treatment recommendations on personal experience rather than on published QOL literature.^65,66,67 Furthermore, this trend could be further explained by the fact that the value of these standardized tools has been questioned by some older clinicians who feel they are able to derive the same information by simply talking to the patients in a less structured way in their clinical practice.^68,69 These results are in contrast to the findings of a study done in Italy which showed that knowledge and use of the HRQoL standardized assessment tools are independent of age, gender, and medical specialty.⁴⁴

Interestingly, radiation oncologists were more likely to assess HRQoL. The reason for this may possibly be because radiation oncologists see the patients more frequently while they are having their radiation treatment and the effect of always asking how the patients are—bringing the patients’ concerns and symptoms to the physicians’ attention.^70,71

Clinical Implications

The findings of this study are very important and can be considered an eye opener to Filipino oncologists. This study showed that although 96% of the respondents reported that HRQoL assessment is important for their clinical work, majority do so verbally using nonstandardized open questions (e.g., How are you?). While this may be of relative benefit to the clinicians, this method of assessment may lead to failure of assessing the full spectrum of HRQoL.^72,73,74 Another important point that Filipino oncologists should realize is that similar to a previous study, physicians sometimes underestimate the impact of the patients’ disease on their HRQoL and therefore should be evaluated using standardized tools.⁷⁵

Limitations

This study had a 40% response rate from Filipino oncologists 312/771. The unequal small sample distribution among the different oncologic subspecialties remains the major limitation of this study probably due to the poor response rate from some oncologic subspecialties.